BERT-INT: A BERT-based Interaction Model For Knowledge Graph Alignment
Xiaobin Tang
1,2
, Jing Zhang
1,2∗
, Bo Chen
1,2
, Yang Yang
3
, Hong Chen
1,2
and Cuiping Li
1,2
1
Key Laboratory of Data Engineering and Knowledge Engineering of Ministry of Education, Renmin
University of China
2
Information School, Renmin University of China
3
Zhejiang University
{txb, zhang-jing, bochen, chong, licuiping}@ruc.edu.cn, yangya@zju.edu.cn
Abstract
Knowledge graph alignment aims to link equivalent
entities across different knowledge graphs. To uti-
lize both the graph structures and the side informa-
tion such as name, description and attributes, most
of the works propagate the side information espe-
cially names through linked entities by graph neural
networks. However, due to the heterogeneity of dif-
ferent knowledge graphs, the alignment accuracy
will be suffered from aggregating different neigh-
bors. This work presents an interaction model to
only leverage the side information. Instead of ag-
gregating neighbors, we compute the interactions
between neighbors which can capture fine-grained
matches of neighbors. Similarly, the interactions of
attributes are also modeled. Experimental results
show that our model significantly outperforms the
best state-of-the-art methods by 1.9-9.7% in terms
of HitRatio@1 on the dataset DBP15K.
1 Introduction
DBpedia, Freebase, YAGO and so on have been published
as noteworthy large and freely available knowledge graphs
(KGs), which can benefit many applications such as question
answering and recommendation
[
Tong et al., 2019
]
. However,
a single KG is far from complete to support such applications
with sufficient facts, which demands an effective way to align
entities across KGs. To solve the problem, much attention has
been paid to leveraging the graph structures to align entities.
The specific technique has evolved from the traditional KG
embedding models such as MTransE
[
Chen et al., 2017
]
and
IPTransE
[
Zhu et al., 2017
]
to recent emergent graph neu-
ral networks such as attention-based GCN
[
Xu et al., 2019
]
,
highway GCN
[
Wu et al., 2019b
]
, relation-aware GCN
[
Wu
et al., 2019a
]
and VR-GCN
[
Ye et al., 2019
]
.
Despite much effort taken on graph structures, the side in-
formation of entities such as name, description and attributes
may play a more important role on many tasks on KGs, be-
cause: (1) the KGs are usually sparse, with a large number
of long-tail entities whose structural embeddings have low
∗
Contact Author
Casino_Royale
_(2006_film)
007
Daniel_Craig
Eva_Green
Columbia_
Pictures
(English)
G
1
G
2
Daniel_
Craig
Eva_Green
Columbia_
Pictures
1.0
1.0
0.0 0.0
0.00.0
0.0 0.0
0.0
Figure 1: A motivating example (Refer to Yang et al., 2019)
expressiveness
[
Guo et al., 2019
]
. For example, in DBpedia-
ZH
1
, the long-tail entities that appear less than 5 times occupy
74.1%; and (2) existing works have demonstrated the effect of
side information on several tasks. For example, HMAN
[
Yang
et al., 2019
]
presents that on the task of entity alignment,
when ignoring entity descriptions, HitRatio@1 will drop at
least 30%.
To leverage the side information to align KGs, the most
popular way is to initialize the embedding of each node by
its side information and apply variant GCN models to up-
date a node embedding by aggregating all neighbors’ embed-
dings
[
Wang et al., 2018; Wu et al., 2019a; Xu et al., 2019
]
.
However, since different KGs are highly heterogeneous, it
is not always the case that equivalent entities share similar
neighbors. Taking an example from HMAN in Figure 1, the
nodes in rectangles are entities to be aligned and the nodes in
ovals are the neighbors. We can see that for the neighbor “En-
glish” in G
2
, no counterpart can be found from the neighbors
in G
1
. In this case, propagating the dissimilar neighbors in
GCNs, especially the hub entities such as the neighbor “En-
glish” with 832 relations, may introduce noises and thus harm
performance
[
Yang et al., 2019
]
. Although some works dis-
tinguish the influence from different neighbors
[
Cao et al.,
2019; Wu et al., 2019a; Xu et al., 2019
]
, essentially, the GCN-
like models still mix the side information of all the neighbors
to represent an entity. HMAN has been aware of the issue,
thus it thoroughly separates the modeling process of side in-
formation and graph structures. However, it discards the side
information of neighbors. Moreover, similar to aggregating
neighbors, it aggregates all the attributes together to repre-
sent an entity like most of the works did
[
Sun et al., 2017;
Zhang et al., 2019
]
, which also results in noisy matches be-
tween entities.
To deal with the noisy matches caused by aggregat-
1
https://wiki.dbpedia.org/downloads-2016-10
Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence (IJCAI-20)
3174
ing neighbors or attributes, we propose BERT-INT—i.e., a
BERT-based INTeraction model to only leverage the side in-
formation, especially the names/descriptions of current en-
tities and neighbors, and attributes of current entities in a
unified way. Generally, we mimic the cognitive process of
comparing two entities by humans, who usually first compare
current entities and then continue to check whether there are
similar neighbors. Following this, based on the embedding
of any piece of name, description and attribute, we compare
each pair of neighbors or attributes, referred to as interac-
tions henceforth, instead of aggregating them. In this way, we
can capture the fine-grained exact/semantic matches between
neighbors and get rid of the negative influence from the dis-
similar neighbors, as presented on the right of Figure 1.
Although we leverage the side information of neighbors,
unlike the variant GCN models, the structural characteristics
of KGs are totally ignored in our model, because we never
propagate the information of neighbors in graphs. From this
point of view, we claim that BERT-INT only leverages side
information. Just because of this, BERT-INT is capable of in-
ductive learning, i.e., we can train BERT-INT on two aligned
KGs and apply it to predict the matches between unseen enti-
ties in other KGs. However, if adopting the variant GCN mod-
els, the test entities should be included in the training KGs.
In summary, we propose an inductive learning-based model
to comprehensively leverage the name/description-view in-
teraction, neighbor-view interactions and attribute-view in-
teractions to align entities. By experiments, we demonstrate
that BERT-INT significantly outperforms the state-of-the-art
models by 1.9-9.7% in HitRatio@1.
2 Problem Definition
Definition 1. Knowledge Graph: We denote a KG as G =
(E, R, A, V ), where each e ∈ E, r ∈ R, a ∈ A and v ∈ V
denotes an entity, a relation, an attribute and a value respec-
tively. N
τ
(e) = {(r
i
, e
i
)}
|N
τ
(e)|
i=1
denotes the set of all the τ-
hop neighbors of entity e with τ as the number of hops, where
each of the i-th neighbor contains a neighboring relation r
i
and the corresponding entity e
i
. A(e) = {(a
i
, v
i
)}
|A(e)|
i=1
de-
notes the set of the attributes of e, where each of the i-th at-
tribute contains a name a
i
and the corresponding value v
i
.
N (e) without τ indicates all the neighbors of e. |N
τ
(e)| and
|A(e)| is the number of N
τ
(e) and A(e) respectively.
Problem 1. Knowledge Graph Alignment: Given two KGs
G and G
′
and a set of already aligned entity pairs I = {(e ∼
e
′
)}, we aim at learning a ranking function f : E × E
′
→ R
to calculate a similarity score between two entities, based on
which we rank the correctly aligned entity e
′
to any queried
entity e as high as possible among all the entities in E
′
.
3 BERT-INT Model
This section introduces the proposed BERT-INT, which con-
sists of a BERT model that is used as a basic represen-
tation unit to embed the name, description, attribute and
value of an entity, and an interaction model built upon the
BERT embeddings to compute the interactions between these
embeddings. The interactions are further divided into the
Concatenation
BERT
Tok NTok 2Tok 1CLS
E
[CLS]
E
1
E
2
E
N
CLS
T
1
T
2
T
N
…
…
…
BERT
Tok NTok 2Tok 1CLS
E
[CLS]
E
1
E
2
E
N
CLS
T
1
T
2
T
N
…
…
…
Pairwise margin loss
Neighbor-view
interactions
Attribute-view
interactions
C(r’
1
)
C(r’
2
)
C(r’
n
)
…
C(e
1
)
C(e
2
)
C(e
n
)
…
C(r
1
)
C(r
2
)
C(r
n
)
…
C(e’
1
)
C(e’
2
)
C(e’
n
)
C(e)
C(e’)
…
Pairwise margin loss
Basic BERT Unit
BERT-based
Interaction Model
MLP
C(v
1
)
C(v
2
)
C(v
n
)
…
C(a
1
)
C(a
2
)
C(a
n
)
…
C(a’
1
)
C(a’
2
)
C(a’
n
)
…
C(v’
1
)
C(v’
2
)
C(v’
n
)
…
Steven
Paul
Apple
Name/description
-view interaction
Neighboring
entity
Neighboring
relation
Attribute
Value
Name/
description
MLP MLP
Figure 2: The framework of BERT-INT. We use BERT as a ba-
sic representation unit to embed the name/description, attribute and
value of an entity, and build neighbor-view/attribute-view interac-
tion models to compute the interactions between these embeddings.
name/description-view interaction, the neighbor-view inter-
actions, and the attribute-view interactions. A unified dual
aggregation function is applied to extract features from the
neighbor-view and attribute-view interactions to estimate
the matching score of entities. Besides, to build compre-
hensive neighbor-view interactions, the interactions between
the neighboring entities, the corresponding neighboring rela-
tions, and the multi-hop neighbors are also considered. Fig-
ure 2 shows the whole framework.
3.1 Basic BERT Unit
We treat entity alignment as the downstream objective to fine-
tune a pre-trained BERT model. Specifically, we first con-
struct the training data D = {(e, e
′+
, e
′−
)}, where each
triplet ( e, e
′+
, e
′−
) ∈ D contains a queried entity e ∈ E, the
rightly aligned counterpart e
′+
∈ E
′
and a negative counter-
part e
′−
randomly sampled from E
′
. For each entity e in the
dataset, we apply a pre-trained multi-lingual BERT
2
to accept
its name/description as the input, filter the CLS embedding of
BERT by a MLP layer to obtain
C(e) = MLP(CLS(e)), (1)
and use a pairwise margin loss to fine-tune BERT:
L =
∑
(e,e
′+
,e
′−
)∈D
max{0, g(e, e
′+
) − g(e, e
′−
) + m)}, (2)
where m is a margin enforced between the positive pairs and
negative pairs, and g(e, e
′
) is instantiated as l
1
distance to
measure the similarity between C(e) and C(e
′
). Negative
pairs are sampled according to the cosine similarity of two
entities
[
Sun et al., 2018
]
.
2
https://github.com/google-research/bert
Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence (IJCAI-20)
3175
For the input of BERT, we give priority to the description as
it contains richer information. We use the name when the de-
scription is missing. Different from HMAN that directly uses
the embeddings of the descriptions to align entities
[
Yang et
al., 2019
]
, we use these embeddings as basic units to compose
the following interaction model. Note we can also connect the
basic BERT unit with the following interaction model as an
end-to-end model to fine-tune BERT and train the interac-
tion model simultaneously. In practice, considering the GPU
memory and the running efficiency, we fine-tune the basic
BERT unit beforehand and freeze its parameters in the fol-
lowing interaction model.
3.2 BERT-based Interaction Model
Based on the basic BERT unit, we build the interaction model
which consists of the name/description-view interaction, the
neighbor-view and the attribute-view interactions.
Name/Description-view Interaction
We apply the basic BERT unit on the names/descriptions of
e and e
′
to obtain C(e) and C(e
′
), and then calculate their
cosine similarity as the name/description-view interaction.
Neighbor-view Interactions
We build the interactions between the neighbors N (e) and
N (e
′
). The general idea is to compare the names/descriptions
of each neighbor pair rather than learning a global represen-
tation for e or e
′
by aggregating the names/descriptions of
all their neighbors as existing works did
[
Wu et al., 2019a;
Xu et al., 2019
]
. This similar idea is widely used in informa-
tion retrieval to capture the exact and soft matches between a
query and a candidate document
[
Xiong et al., 2017
]
. Specif-
ically, we apply the basic BERT unit to obtain {C(e
i
)}
|N (e)|
i=1
and {C(e
′
i
)}
|N (e
′
)|
i=1
for e and e
′
’s neighboring entities based
on their names/descriptions, compute a similarity matrix be-
tween the two embedding sets and then apply a dual aggrega-
tion function to extract the similarity features from the matrix.
We use S to represent the interactions between the neighbors
of e and e
′
, with each element s
ij
standing for the interac-
tion, i.e., the cosine similarity s
ij
=
C(e
i
)·C(e
′
j
)
∥C(e
i
)∥·∥C(e
′
j
)∥
between
C(e
i
) and C(e
′
j
), where C(e
i
) and C(e
′
j
) are obtained by
Eq.(1) for the i-th neighbor of e and the j-th neighbor of e
′
respectively.
We apply a dual aggregation function to extract the similar-
ity features along both the rows and columns of S. CNN and
RNN are usually used as the aggregation function to extract
the matching patterns from the similarity matrix between two
sentences. Different from sentences, the neighbors are dis-
ordered and independent from each other. Thus we adopt a
RBF kernel aggregation function
[
Xiong et al., 2017
]
to ex-
tract features about the accumulation of similarities.
Before making use of the RBF kernel aggregations, we
first apply a max pooling operation on each row S
i
=
{s
i0
, · · · , s
ij
, · · · , s
in
} to get the maximal similarity s
max
i
,
i.e., we select the most possibly aligned counterpart from
the neighbors of e
′
for the i-th neighbor of e. This is be-
cause, following the one-one mapping assumption, we only
care about how similar the most possibly aligned counterpart
C(e’
1
)
C(e’
2
)
C(e’
n
)
…
C(e
1
)
C(e
2
)
C(e
n
)
…
Average
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
Neighboring entity similarity matrix
s
00
<latexit sha1_base64="liuNhAErO5iCfohbJ8KP/dpmJ/Y=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aoZZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/rz48s</latexit>
<latexit sha1_base64="liuNhAErO5iCfohbJ8KP/dpmJ/Y=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aoZZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/rz48s</latexit>
<latexit sha1_base64="liuNhAErO5iCfohbJ8KP/dpmJ/Y=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aoZZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/rz48s</latexit>
<latexit sha1_base64="liuNhAErO5iCfohbJ8KP/dpmJ/Y=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aoZZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/rz48s</latexit>
s
01
<latexit sha1_base64="t8XVwIeWyQN1j8wid15WWVkfm0w=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzPVm1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVY8t</latexit>
<latexit sha1_base64="t8XVwIeWyQN1j8wid15WWVkfm0w=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzPVm1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVY8t</latexit>
<latexit sha1_base64="t8XVwIeWyQN1j8wid15WWVkfm0w=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzPVm1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVY8t</latexit>
<latexit sha1_base64="t8XVwIeWyQN1j8wid15WWVkfm0w=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzPVm1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVY8t</latexit>
s
0n
<latexit sha1_base64="8EIBYynhbFV80R8DzKcjgpT/+Jc=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzJWz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KUo9q</latexit>
<latexit sha1_base64="8EIBYynhbFV80R8DzKcjgpT/+Jc=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzJWz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KUo9q</latexit>
<latexit sha1_base64="8EIBYynhbFV80R8DzKcjgpT/+Jc=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzJWz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KUo9q</latexit>
<latexit sha1_base64="8EIBYynhbFV80R8DzKcjgpT/+Jc=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzJWz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KUo9q</latexit>
s
10
<latexit sha1_base64="B4RV+mxWdhh1b3BE0G/iT8BAANM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzHNn1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVo8t</latexit>
<latexit sha1_base64="B4RV+mxWdhh1b3BE0G/iT8BAANM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzHNn1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVo8t</latexit>
<latexit sha1_base64="B4RV+mxWdhh1b3BE0G/iT8BAANM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzHNn1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVo8t</latexit>
<latexit sha1_base64="B4RV+mxWdhh1b3BE0G/iT8BAANM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDzHNn1UG54tbcBcg68XJSgRzNQfmrP4xZGqE0TFCte56bGD+jynAmcFbqpxoTyiZ0hD1LJY1Q+9ni3hm5sMqQhLGyJQ1ZqL8nMhppPY0C2xlRM9ar3lz8z+ulJrzxMy6T1KBky0VhKoiJyfx5MuQKmRFTSyhT3N5K2JgqyoyNqGRD8FZfXiftes1za959vdK4yuMowhmcwyV4cA0NuIMmtICBgGd4hTfn0Xlx3p2PZWvByWdO4Q+czx/tVo8t</latexit>
s
11
<latexit sha1_base64="EVXSLjBwYM6ovQiZ1udEYh4ijOU=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQbLImWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNirVlLWoEkp3Q2KY4JK1LLeCdRPNSBwK1gknt3O/88S04Uo+2GnCgpiMJI84JdZJ3aoZZBjPqoNyxa/5C6B1gnNSgRzNQfmrP1Q0jZm0VBBjethPbJARbTkVbFbqp4YlhE7IiPUclSRmJsgW987QhVOGKFLalbRoof6eyEhszDQOXWdM7NisenPxP6+X2ugmyLhMUsskXS6KUoGsQvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/u3I8u</latexit>
<latexit sha1_base64="EVXSLjBwYM6ovQiZ1udEYh4ijOU=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQbLImWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNirVlLWoEkp3Q2KY4JK1LLeCdRPNSBwK1gknt3O/88S04Uo+2GnCgpiMJI84JdZJ3aoZZBjPqoNyxa/5C6B1gnNSgRzNQfmrP1Q0jZm0VBBjethPbJARbTkVbFbqp4YlhE7IiPUclSRmJsgW987QhVOGKFLalbRoof6eyEhszDQOXWdM7NisenPxP6+X2ugmyLhMUsskXS6KUoGsQvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/u3I8u</latexit>
<latexit sha1_base64="EVXSLjBwYM6ovQiZ1udEYh4ijOU=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQbLImWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNirVlLWoEkp3Q2KY4JK1LLeCdRPNSBwK1gknt3O/88S04Uo+2GnCgpiMJI84JdZJ3aoZZBjPqoNyxa/5C6B1gnNSgRzNQfmrP1Q0jZm0VBBjethPbJARbTkVbFbqp4YlhE7IiPUclSRmJsgW987QhVOGKFLalbRoof6eyEhszDQOXWdM7NisenPxP6+X2ugmyLhMUsskXS6KUoGsQvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/u3I8u</latexit>
<latexit sha1_base64="EVXSLjBwYM6ovQiZ1udEYh4ijOU=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQbLImWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNirVlLWoEkp3Q2KY4JK1LLeCdRPNSBwK1gknt3O/88S04Uo+2GnCgpiMJI84JdZJ3aoZZBjPqoNyxa/5C6B1gnNSgRzNQfmrP1Q0jZm0VBBjethPbJARbTkVbFbqp4YlhE7IiPUclSRmJsgW987QhVOGKFLalbRoof6eyEhszDQOXWdM7NisenPxP6+X2ugmyLhMUsskXS6KUoGsQvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/u3I8u</latexit>
s
1n
<latexit sha1_base64="3ooisHfUTyVs7d56tOnsxcXiQZQ=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZJ6aVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9L2Y9r</latexit>
<latexit sha1_base64="3ooisHfUTyVs7d56tOnsxcXiQZQ=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZJ6aVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9L2Y9r</latexit>
<latexit sha1_base64="3ooisHfUTyVs7d56tOnsxcXiQZQ=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZJ6aVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9L2Y9r</latexit>
<latexit sha1_base64="3ooisHfUTyVs7d56tOnsxcXiQZQ=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZJ6aVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9L2Y9r</latexit>
s
n0
<latexit sha1_base64="RqHwy1IVbuXsU/iBgFKx6bAd5fs=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTLqz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KkI9q</latexit>
<latexit sha1_base64="RqHwy1IVbuXsU/iBgFKx6bAd5fs=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTLqz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KkI9q</latexit>
<latexit sha1_base64="RqHwy1IVbuXsU/iBgFKx6bAd5fs=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTLqz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KkI9q</latexit>
<latexit sha1_base64="RqHwy1IVbuXsU/iBgFKx6bAd5fs=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTLqz6qBccWvuAmSdeDmpQI7moPzVH8YsjVAaJqjWPc9NjJ9RZTgTOCv1U40JZRM6wp6lkkao/Wxx74xcWGVIwljZkoYs1N8TGY20nkaB7YyoGetVby7+5/VSE974GZdJalCy5aIwFcTEZP48GXKFzIipJZQpbm8lbEwVZcZGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwDO8wpvz6Lw4787HsrXg5DOn8AfO5w9KkI9q</latexit>
s
n1
<latexit sha1_base64="087dM6uzBMXa+FOcpSWEuUIousM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZMqbVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9MFo9r</latexit>
<latexit sha1_base64="087dM6uzBMXa+FOcpSWEuUIousM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZMqbVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9MFo9r</latexit>
<latexit sha1_base64="087dM6uzBMXa+FOcpSWEuUIousM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZMqbVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9MFo9r</latexit>
<latexit sha1_base64="087dM6uzBMXa+FOcpSWEuUIousM=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsN+3SzSbuToQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IJHCoOt+O4WNza3tneJuaW//4PCofHzSNnGqGW+xWMa6G1DDpVC8hQIl7yaa0yiQvBNMbud+54lrI2L1gNOE+xEdKREKRtFK3aoZZMqbVQfliltzFyDrxMtJBXI0B+Wv/jBmacQVMkmN6Xlugn5GNQom+azUTw1PKJvQEe9ZqmjEjZ8t7p2RC6sMSRhrWwrJQv09kdHImGkU2M6I4tisenPxP6+XYnjjZ0IlKXLFlovCVBKMyfx5MhSaM5RTSyjTwt5K2JhqytBGVLIheKsvr5N2vea5Ne++Xmlc5XEU4QzO4RI8uIYG3EETWsBAwjO8wpvz6Lw4787HsrXg5DOn8AfO5w9MFo9r</latexit>
s
nn
<latexit sha1_base64="mwh91s92BV8IaqzAC9g98k6K4A4=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTMpZdVCuuDV3AbJOvJxUIEdzUP7qD2OWRigNE1Trnucmxs+oMpwJnJX6qcaEsgkdYc9SSSPUfra4d0YurDIkYaxsSUMW6u+JjEZaT6PAdkbUjPWqNxf/83qpCW/8jMskNSjZclGYCmJiMn+eDLlCZsTUEsoUt7cSNqaKMmMjKtkQvNWX10m7XvPcmndfrzSu8jiKcAbncAkeXEMD7qAJLWAg4Ble4c15dF6cd+dj2Vpw8plT+APn8wepBI+o</latexit>
<latexit sha1_base64="mwh91s92BV8IaqzAC9g98k6K4A4=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTMpZdVCuuDV3AbJOvJxUIEdzUP7qD2OWRigNE1Trnucmxs+oMpwJnJX6qcaEsgkdYc9SSSPUfra4d0YurDIkYaxsSUMW6u+JjEZaT6PAdkbUjPWqNxf/83qpCW/8jMskNSjZclGYCmJiMn+eDLlCZsTUEsoUt7cSNqaKMmMjKtkQvNWX10m7XvPcmndfrzSu8jiKcAbncAkeXEMD7qAJLWAg4Ble4c15dF6cd+dj2Vpw8plT+APn8wepBI+o</latexit>
<latexit sha1_base64="mwh91s92BV8IaqzAC9g98k6K4A4=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTMpZdVCuuDV3AbJOvJxUIEdzUP7qD2OWRigNE1Trnucmxs+oMpwJnJX6qcaEsgkdYc9SSSPUfra4d0YurDIkYaxsSUMW6u+JjEZaT6PAdkbUjPWqNxf/83qpCW/8jMskNSjZclGYCmJiMn+eDLlCZsTUEsoUt7cSNqaKMmMjKtkQvNWX10m7XvPcmndfrzSu8jiKcAbncAkeXEMD7qAJLWAg4Ble4c15dF6cd+dj2Vpw8plT+APn8wepBI+o</latexit>
<latexit sha1_base64="mwh91s92BV8IaqzAC9g98k6K4A4=">AAAB73icbVBNS8NAEJ3Ur1q/qh69LLaCp5IUQY8FLx4r2A9oQ9lsJ+3SzSbuboQS+ie8eFDEq3/Hm//GbZuDtj4YeLw3w8y8IBFcG9f9dgobm1vbO8Xd0t7+weFR+fikreNUMWyxWMSqG1CNgktsGW4EdhOFNAoEdoLJ7dzvPKHSPJYPZpqgH9GR5CFn1FipW9WDTMpZdVCuuDV3AbJOvJxUIEdzUP7qD2OWRigNE1Trnucmxs+oMpwJnJX6qcaEsgkdYc9SSSPUfra4d0YurDIkYaxsSUMW6u+JjEZaT6PAdkbUjPWqNxf/83qpCW/8jMskNSjZclGYCmJiMn+eDLlCZsTUEsoUt7cSNqaKMmMjKtkQvNWX10m7XvPcmndfrzSu8jiKcAbncAkeXEMD7qAJLWAg4Ble4c15dF6cd+dj2Vpw8plT+APn8wepBI+o</latexit>
Neighboring relation mask matrix
C(r
1
)
C(r
2
)
C(r
n
)
…
C(r’
1
)
C(r’
2
)
C(r’
n
)
…
⌦
<latexit sha1_base64="ywicrLnvnDH78C723eeA8/jhO+M=">AAAB8HicbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9JljA7mU2GzGOZmRXCkq/w4kERr36ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk56rPaVZYKa6qBc8Wv+AmidBDmpQI7moPzVHyqSCiot4diYXuAnNsywtoxwOiv1U0MTTCZ4RHuOSuy2hNni4Bm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzb9HQ6YpsXzqCCaauVsRGWONiXUZlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhAQ8Ayv8OZp78V79z6WrQUvnzmFP/A+fwA8M4/5</latexit>
<latexit sha1_base64="ywicrLnvnDH78C723eeA8/jhO+M=">AAAB8HicbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9JljA7mU2GzGOZmRXCkq/w4kERr36ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk56rPaVZYKa6qBc8Wv+AmidBDmpQI7moPzVHyqSCiot4diYXuAnNsywtoxwOiv1U0MTTCZ4RHuOSuy2hNni4Bm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzb9HQ6YpsXzqCCaauVsRGWONiXUZlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhAQ8Ayv8OZp78V79z6WrQUvnzmFP/A+fwA8M4/5</latexit>
<latexit sha1_base64="ywicrLnvnDH78C723eeA8/jhO+M=">AAAB8HicbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9JljA7mU2GzGOZmRXCkq/w4kERr36ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk56rPaVZYKa6qBc8Wv+AmidBDmpQI7moPzVHyqSCiot4diYXuAnNsywtoxwOiv1U0MTTCZ4RHuOSuy2hNni4Bm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzb9HQ6YpsXzqCCaauVsRGWONiXUZlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhAQ8Ayv8OZp78V79z6WrQUvnzmFP/A+fwA8M4/5</latexit>
<latexit sha1_base64="ywicrLnvnDH78C723eeA8/jhO+M=">AAAB8HicbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9JljA7mU2GzGOZmRXCkq/w4kERr36ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk56rPaVZYKa6qBc8Wv+AmidBDmpQI7moPzVHyqSCiot4diYXuAnNsywtoxwOiv1U0MTTCZ4RHuOSuy2hNni4Bm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzb9HQ6YpsXzqCCaauVsRGWONiXUZlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhAQ8Ayv8OZp78V79z6WrQUvnzmFP/A+fwA8M4/5</latexit>
c
(N (e), N (e
0
))
<latexit sha1_base64="0tN/Wtf93mJp8sQ3/fOmdpCYgKM=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN5jyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX78flJvZ</latexit>
<latexit sha1_base64="0tN/Wtf93mJp8sQ3/fOmdpCYgKM=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN5jyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX78flJvZ</latexit>
<latexit sha1_base64="0tN/Wtf93mJp8sQ3/fOmdpCYgKM=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN5jyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX78flJvZ</latexit>
<latexit sha1_base64="0tN/Wtf93mJp8sQ3/fOmdpCYgKM=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN5jyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX78flJvZ</latexit>
m
00
<latexit sha1_base64="77lo0QiQd7YmxmbO6o8BG45bpmQ=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aocZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/ik48m</latexit>
<latexit sha1_base64="77lo0QiQd7YmxmbO6o8BG45bpmQ=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aocZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/ik48m</latexit>
<latexit sha1_base64="77lo0QiQd7YmxmbO6o8BG45bpmQ=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aocZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/ik48m</latexit>
<latexit sha1_base64="77lo0QiQd7YmxmbO6o8BG45bpmQ=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp5Itgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhuLMbfXmFjc2t7p7hb2ts/ODwqH5+0TZxqylo0FrHuhsQwwRVrWW4F6yaaERkK1gknt3O/88S04bF6sNOEBZKMFI84JdZJ3aocZBjPqoNyBdfwAmid+DmpQI7moPzVH8Y0lUxZKogxPR8nNsiItpwKNiv1U8MSQidkxHqOKiKZCbLFvTN04ZQhimLtSlm0UH9PZEQaM5Wh65TEjs2qNxf/83qpjW6CjKsktUzR5aIoFcjGaP48GnLNqBVTRwjV3N2K6JhoQq2LqORC8FdfXiftes3HNf++Xmlc5XEU4QzO4RJ8uIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/ik48m</latexit>
m
01
<latexit sha1_base64="SXsV6h/Yn1VclffdBJkOYNweLuc=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg84NZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGY8n</latexit>
<latexit sha1_base64="SXsV6h/Yn1VclffdBJkOYNweLuc=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg84NZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGY8n</latexit>
<latexit sha1_base64="SXsV6h/Yn1VclffdBJkOYNweLuc=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg84NZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGY8n</latexit>
<latexit sha1_base64="SXsV6h/Yn1VclffdBJkOYNweLuc=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg84NZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGY8n</latexit>
m
0n
<latexit sha1_base64="f1w9NiTfZdWA6BhUlkk+nUQaoVI=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg8+WsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBFo9k</latexit>
<latexit sha1_base64="f1w9NiTfZdWA6BhUlkk+nUQaoVI=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg8+WsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBFo9k</latexit>
<latexit sha1_base64="f1w9NiTfZdWA6BhUlkk+nUQaoVI=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg8+WsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBFo9k</latexit>
<latexit sha1_base64="f1w9NiTfZdWA6BhUlkk+nUQaoVI=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUg8+WsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBFo9k</latexit>
m
10
<latexit sha1_base64="OrG4nvGz6KHerv62wV5IYAGEKcM=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC/xZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGo8n</latexit>
<latexit sha1_base64="OrG4nvGz6KHerv62wV5IYAGEKcM=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC/xZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGo8n</latexit>
<latexit sha1_base64="OrG4nvGz6KHerv62wV5IYAGEKcM=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC/xZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGo8n</latexit>
<latexit sha1_base64="OrG4nvGz6KHerv62wV5IYAGEKcM=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC/xZdVCu+DV/AbROgpxUIEdzUP7qDxVJBZWWcGxML/ATG2ZYW0Y4nZX6qaEJJhM8oj1HJRbUhNni3hm6cMoQxUq7khYt1N8TGRbGTEXkOgW2Y7PqzcX/vF5q45swYzJJLZVkuShOObIKzZ9HQ6YpsXzqCCaauVsRGWONiXURlVwIwerL66RdrwV+LbivVxpXeRxFOINzuIQArqEBd9CEFhDg8Ayv8OY9ei/eu/exbC14+cwp/IH3+QPkGo8n</latexit>
m
11
<latexit sha1_base64="E7jg/k6Hi8I6N1dtD1n3xxoXvZM=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNnGqKWvRWMS6GxLDBFesZbkVrJtoRmQoWCec3M79zhPThsfqwU4TFkgyUjzilFgndatykGE8qw7KFb/mL4DWCc5JBXI0B+Wv/jCmqWTKUkGM6WE/sUFGtOVUsFmpnxqWEDohI9ZzVBHJTJAt7p2hC6cMURRrV8qihfp7IiPSmKkMXackdmxWvbn4n9dLbXQTZFwlqWWKLhdFqUA2RvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/loI8o</latexit>
<latexit sha1_base64="E7jg/k6Hi8I6N1dtD1n3xxoXvZM=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNnGqKWvRWMS6GxLDBFesZbkVrJtoRmQoWCec3M79zhPThsfqwU4TFkgyUjzilFgndatykGE8qw7KFb/mL4DWCc5JBXI0B+Wv/jCmqWTKUkGM6WE/sUFGtOVUsFmpnxqWEDohI9ZzVBHJTJAt7p2hC6cMURRrV8qihfp7IiPSmKkMXackdmxWvbn4n9dLbXQTZFwlqWWKLhdFqUA2RvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/loI8o</latexit>
<latexit sha1_base64="E7jg/k6Hi8I6N1dtD1n3xxoXvZM=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNnGqKWvRWMS6GxLDBFesZbkVrJtoRmQoWCec3M79zhPThsfqwU4TFkgyUjzilFgndatykGE8qw7KFb/mL4DWCc5JBXI0B+Wv/jCmqWTKUkGM6WE/sUFGtOVUsFmpnxqWEDohI9ZzVBHJTJAt7p2hC6cMURRrV8qihfp7IiPSmKkMXackdmxWvbn4n9dLbXQTZFwlqWWKLhdFqUA2RvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/loI8o</latexit>
<latexit sha1_base64="E7jg/k6Hi8I6N1dtD1n3xxoXvZM=">AAAB73icbVBNSwMxEJ2tX7V+VT16CbaCp7Ipgh4LXjxWsB/QLiWbZtvQJLsmWaEs/RNePCji1b/jzX9j2u5BWx8MPN6bYWZemAhurO9/e4WNza3tneJuaW//4PCofHzSNnGqKWvRWMS6GxLDBFesZbkVrJtoRmQoWCec3M79zhPThsfqwU4TFkgyUjzilFgndatykGE8qw7KFb/mL4DWCc5JBXI0B+Wv/jCmqWTKUkGM6WE/sUFGtOVUsFmpnxqWEDohI9ZzVBHJTJAt7p2hC6cMURRrV8qihfp7IiPSmKkMXackdmxWvbn4n9dLbXQTZFwlqWWKLhdFqUA2RvPn0ZBrRq2YOkKo5u5WRMdEE2pdRCUXAl59eZ206zXs1/B9vdK4yuMowhmcwyVguIYG3EETWkBBwDO8wpv36L14797HsrXg5TOn8Afe5w/loI8o</latexit>
m
1n
<latexit sha1_base64="T2OSxE+/4QDZkl4BuCZ2Lbey0Iw=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC+SsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFCnY9l</latexit>
<latexit sha1_base64="T2OSxE+/4QDZkl4BuCZ2Lbey0Iw=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC+SsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFCnY9l</latexit>
<latexit sha1_base64="T2OSxE+/4QDZkl4BuCZ2Lbey0Iw=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC+SsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFCnY9l</latexit>
<latexit sha1_base64="T2OSxE+/4QDZkl4BuCZ2Lbey0Iw=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgC+SsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFCnY9l</latexit>
m
n0
<latexit sha1_base64="82Y5nLVAo4+sFCvnPscI+y+4c9w=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk/6sOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBVI9k</latexit>
<latexit sha1_base64="82Y5nLVAo4+sFCvnPscI+y+4c9w=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk/6sOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBVI9k</latexit>
<latexit sha1_base64="82Y5nLVAo4+sFCvnPscI+y+4c9w=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk/6sOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBVI9k</latexit>
<latexit sha1_base64="82Y5nLVAo4+sFCvnPscI+y+4c9w=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk/6sOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFBVI9k</latexit>
m
n1
<latexit sha1_base64="eHNjwaymQNOUa5D5CZK5p1SDE7s=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk8GsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFC2o9l</latexit>
<latexit sha1_base64="eHNjwaymQNOUa5D5CZK5p1SDE7s=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk8GsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFC2o9l</latexit>
<latexit sha1_base64="eHNjwaymQNOUa5D5CZK5p1SDE7s=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk8GsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFC2o9l</latexit>
<latexit sha1_base64="eHNjwaymQNOUa5D5CZK5p1SDE7s=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk8GsOihX/Jq/AFonQU4qkKM5KH/1h4qkgkpLODamF/iJDTOsLSOczkr91NAEkwke0Z6jEgtqwmxx7wxdOGWIYqVdSYsW6u+JDAtjpiJynQLbsVn15uJ/Xi+18U2YMZmklkqyXBSnHFmF5s+jIdOUWD51BBPN3K2IjLHGxLqISi6EYPXlddKu1wK/FtzXK42rPI4inME5XEIA19CAO2hCCwhweIZXePMevRfv3ftYtha8fOYU/sD7/AFC2o9l</latexit>
m
nn
<latexit sha1_base64="daEkMfU2q7P3lrQsEeska5ChWkE=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk3JWHZQrfs1fAK2TICcVyNEclL/6Q0VSQaUlHBvTC/zEhhnWlhFOZ6V+amiCyQSPaM9RiQU1Yba4d4YunDJEsdKupEUL9fdEhoUxUxG5ToHt2Kx6c/E/r5fa+CbMmExSSyVZLopTjqxC8+fRkGlKLJ86golm7lZExlhjYl1EJRdCsPryOmnXa4FfC+7rlcZVHkcRzuAcLiGAa2jAHTShBQQ4PMMrvHmP3ov37n0sWwtePnMKf+B9/gCfyI+i</latexit>
<latexit sha1_base64="daEkMfU2q7P3lrQsEeska5ChWkE=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk3JWHZQrfs1fAK2TICcVyNEclL/6Q0VSQaUlHBvTC/zEhhnWlhFOZ6V+amiCyQSPaM9RiQU1Yba4d4YunDJEsdKupEUL9fdEhoUxUxG5ToHt2Kx6c/E/r5fa+CbMmExSSyVZLopTjqxC8+fRkGlKLJ86golm7lZExlhjYl1EJRdCsPryOmnXa4FfC+7rlcZVHkcRzuAcLiGAa2jAHTShBQQ4PMMrvHmP3ov37n0sWwtePnMKf+B9/gCfyI+i</latexit>
<latexit sha1_base64="daEkMfU2q7P3lrQsEeska5ChWkE=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk3JWHZQrfs1fAK2TICcVyNEclL/6Q0VSQaUlHBvTC/zEhhnWlhFOZ6V+amiCyQSPaM9RiQU1Yba4d4YunDJEsdKupEUL9fdEhoUxUxG5ToHt2Kx6c/E/r5fa+CbMmExSSyVZLopTjqxC8+fRkGlKLJ86golm7lZExlhjYl1EJRdCsPryOmnXa4FfC+7rlcZVHkcRzuAcLiGAa2jAHTShBQQ4PMMrvHmP3ov37n0sWwtePnMKf+B9/gCfyI+i</latexit>
<latexit sha1_base64="daEkMfU2q7P3lrQsEeska5ChWkE=">AAAB73icbVDLSgNBEOyNrxhfUY9eBhPBU9gNgh4DXjxGMA9IljA7mU2GzGOdmRXCkp/w4kERr/6ON//GSbIHTSxoKKq66e6KEs6M9f1vr7CxubW9U9wt7e0fHB6Vj0/aRqWa0BZRXOluhA3lTNKWZZbTbqIpFhGnnWhyO/c7T1QbpuSDnSY0FHgkWcwItk7qVsUgk3JWHZQrfs1fAK2TICcVyNEclL/6Q0VSQaUlHBvTC/zEhhnWlhFOZ6V+amiCyQSPaM9RiQU1Yba4d4YunDJEsdKupEUL9fdEhoUxUxG5ToHt2Kx6c/E/r5fa+CbMmExSSyVZLopTjqxC8+fRkGlKLJ86golm7lZExlhjYl1EJRdCsPryOmnXa4FfC+7rlcZVHkcRzuAcLiGAa2jAHTShBQQ4PMMrvHmP3ov37n0sWwtePnMKf+B9/gCfyI+i</latexit>
Max-
pooling
Max-
pooling
K
r
(S
1
)
<latexit sha1_base64="ZqdFLXt7NWs9I4TBTQLAQyYstTk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fd5iX1g==</latexit>
<latexit sha1_base64="ZqdFLXt7NWs9I4TBTQLAQyYstTk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fd5iX1g==</latexit>
<latexit sha1_base64="ZqdFLXt7NWs9I4TBTQLAQyYstTk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fd5iX1g==</latexit>
<latexit sha1_base64="ZqdFLXt7NWs9I4TBTQLAQyYstTk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fd5iX1g==</latexit>
Kernel
aggregation
Kernel
aggregation
Max-pooling
Max-pooling
Kernel aggregation
Kernel aggregation
K
c
(S
n
)
<latexit sha1_base64="JP8jiEfTdSj4wGtu3SjUyTtPShY=">AAACB3icbVDLSsNAFJ3UV62vqEtBBluhbkpSBF0W3AhuKtoHtDFMppN26GQSZiZCCdm58VfcuFDErb/gzr9x0kbQ1gMDZ865l3vv8SJGpbKsL6OwtLyyulZcL21sbm3vmLt7bRnGApMWDlkouh6ShFFOWooqRrqRICjwGOl444vM79wTIWnIb9UkIk6Ahpz6FCOlJdc8rPQDpEaen1yld7j687lJ3YSnJxXXLFs1awq4SOyclEGOpmt+9gchjgPCFWZIyp5tRcpJkFAUM5KW+rEkEcJjNCQ9TTkKiHSS6R0pPNbKAPqh0I8rOFV/dyQokHISeLoy21POe5n4n9eLlX/uJJRHsSIczwb5MYMqhFkocEAFwYpNNEFYUL0rxCMkEFY6upIOwZ4/eZG06zXbqtnX9XLjNI+jCA7AEagCG5yBBrgETdACGDyAJ/ACXo1H49l4M95npQUj79kHf2B8fAOV25kQ</latexit>
<latexit sha1_base64="JP8jiEfTdSj4wGtu3SjUyTtPShY=">AAACB3icbVDLSsNAFJ3UV62vqEtBBluhbkpSBF0W3AhuKtoHtDFMppN26GQSZiZCCdm58VfcuFDErb/gzr9x0kbQ1gMDZ865l3vv8SJGpbKsL6OwtLyyulZcL21sbm3vmLt7bRnGApMWDlkouh6ShFFOWooqRrqRICjwGOl444vM79wTIWnIb9UkIk6Ahpz6FCOlJdc8rPQDpEaen1yld7j687lJ3YSnJxXXLFs1awq4SOyclEGOpmt+9gchjgPCFWZIyp5tRcpJkFAUM5KW+rEkEcJjNCQ9TTkKiHSS6R0pPNbKAPqh0I8rOFV/dyQokHISeLoy21POe5n4n9eLlX/uJJRHsSIczwb5MYMqhFkocEAFwYpNNEFYUL0rxCMkEFY6upIOwZ4/eZG06zXbqtnX9XLjNI+jCA7AEagCG5yBBrgETdACGDyAJ/ACXo1H49l4M95npQUj79kHf2B8fAOV25kQ</latexit>
<latexit sha1_base64="JP8jiEfTdSj4wGtu3SjUyTtPShY=">AAACB3icbVDLSsNAFJ3UV62vqEtBBluhbkpSBF0W3AhuKtoHtDFMppN26GQSZiZCCdm58VfcuFDErb/gzr9x0kbQ1gMDZ865l3vv8SJGpbKsL6OwtLyyulZcL21sbm3vmLt7bRnGApMWDlkouh6ShFFOWooqRrqRICjwGOl444vM79wTIWnIb9UkIk6Ahpz6FCOlJdc8rPQDpEaen1yld7j687lJ3YSnJxXXLFs1awq4SOyclEGOpmt+9gchjgPCFWZIyp5tRcpJkFAUM5KW+rEkEcJjNCQ9TTkKiHSS6R0pPNbKAPqh0I8rOFV/dyQokHISeLoy21POe5n4n9eLlX/uJJRHsSIczwb5MYMqhFkocEAFwYpNNEFYUL0rxCMkEFY6upIOwZ4/eZG06zXbqtnX9XLjNI+jCA7AEagCG5yBBrgETdACGDyAJ/ACXo1H49l4M95npQUj79kHf2B8fAOV25kQ</latexit>
<latexit sha1_base64="JP8jiEfTdSj4wGtu3SjUyTtPShY=">AAACB3icbVDLSsNAFJ3UV62vqEtBBluhbkpSBF0W3AhuKtoHtDFMppN26GQSZiZCCdm58VfcuFDErb/gzr9x0kbQ1gMDZ865l3vv8SJGpbKsL6OwtLyyulZcL21sbm3vmLt7bRnGApMWDlkouh6ShFFOWooqRrqRICjwGOl444vM79wTIWnIb9UkIk6Ahpz6FCOlJdc8rPQDpEaen1yld7j687lJ3YSnJxXXLFs1awq4SOyclEGOpmt+9gchjgPCFWZIyp5tRcpJkFAUM5KW+rEkEcJjNCQ9TTkKiHSS6R0pPNbKAPqh0I8rOFV/dyQokHISeLoy21POe5n4n9eLlX/uJJRHsSIczwb5MYMqhFkocEAFwYpNNEFYUL0rxCMkEFY6upIOwZ4/eZG06zXbqtnX9XLjNI+jCA7AEagCG5yBBrgETdACGDyAJ/ACXo1H49l4M95npQUj79kHf2B8fAOV25kQ</latexit>
K
c
(S
1
)
<latexit sha1_base64="D8M0uvM2U7oXjOrNkSjUveE+Nqk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK73D153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fX/uXxw==</latexit>
<latexit sha1_base64="D8M0uvM2U7oXjOrNkSjUveE+Nqk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK73D153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fX/uXxw==</latexit>
<latexit sha1_base64="D8M0uvM2U7oXjOrNkSjUveE+Nqk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK73D153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fX/uXxw==</latexit>
<latexit sha1_base64="D8M0uvM2U7oXjOrNkSjUveE+Nqk=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK73D153OTuvZJxTXLVs2aAi4SOydlkKPpmp/9QYjjgHCFGZKyZ1uRchIkFMWMpKV+LEmE8BgNSU9TjgIinWR6RQqPtTKAfij04wpO1d8dCQqknASersy2lPNeJv7n9WLlnzsJ5VGsCMezQX7MoAphFgkcUEGwYhNNEBZU7wrxCAmElQ6upEOw509eJO16zbZq9nW93DjN4yiCA3AEqsAGZ6ABLkETtAAGD+AJvIBX49F4Nt6M91lpwch79sEfGB/fX/uXxw==</latexit>
…
…
Concatenation
Average
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
<latexit sha1_base64="0OUc/JQHkXv7pnGl3fJcspbxSs8=">AAAB+XicbVC7TsMwFL0pr1JeAUaWiBaJqUq6wFgEA2OR6ENqo8pxndaq40T2TUUV9U9YGECIlT9h429w2wzQciRLR+eca1+fIBFco+t+W4WNza3tneJuaW//4PDIPj5p6ThVlDVpLGLVCYhmgkvWRI6CdRLFSBQI1g7Gt3O/PWFK81g+4jRhfkSGkoecEjRS37YrPWRPmN2YEBmyWaVvl92qu4CzTryclCFHo29/9QYxTSMmkQqidddzE/QzopBTwWalXqpZQujY3N41VJKIaT9bbD5zLowycMJYmSPRWai/JzISaT2NApOMCI70qjcX//O6KYbXfsZlkiKTdPlQmAoHY2degzPgilEUU0MIVdzs6tARUYSiKatkSvBWv7xOWrWq51a9h1q5fpfXUYQzOIdL8OAK6nAPDWgChQk8wyu8WZn1Yr1bH8towcpnTuEPrM8fQxqTZw==</latexit>
r
(N (e), N (e
0
))
<latexit sha1_base64="zF0xY2+k5urDiLzGTDOifoDPWGk=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN4Lyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX784P5vo</latexit>
<latexit sha1_base64="zF0xY2+k5urDiLzGTDOifoDPWGk=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN4Lyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX784P5vo</latexit>
<latexit sha1_base64="zF0xY2+k5urDiLzGTDOifoDPWGk=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN4Lyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX784P5vo</latexit>
<latexit sha1_base64="zF0xY2+k5urDiLzGTDOifoDPWGk=">AAACEXicbVDLSsNAFJ3UV62vqEs3g62YgpSkCLosuHElFewDmlgm00k7OJmEmYlQQn/Bjb/ixoUibt2582+ctFnY1gMDZ865l3vv8WNGpbLtH6Owsrq2vlHcLG1t7+zumfsHbRklApMWjlgkuj6ShFFOWooqRrqxICj0Gen4D1eZ33kkQtKI36lxTLwQDTkNKEZKS33TqrjxiN4Lyw2RGmHE0puJRapnc9/TarXSN8t2zZ4CLhMnJ2WQo9k3v91BhJOQcIUZkrLn2LHyUiQUxYxMSm4iSYzwAxqSnqYchUR66fSiCTzRygAGkdCPKzhV/3akKJRyHPq6MltULnqZ+J/XS1Rw6aWUx4kiHM8GBQmDKoJZPHBABcGKjTVBWFC9K8QjJBBWOsSSDsFZPHmZtOs1x645t/Vy4zyPowiOwDGwgAMuQANcgyZoAQyewAt4A+/Gs/FqfBifs9KCkfccgjkYX784P5vo</latexit>
(N (e), N (e
0
))
<latexit sha1_base64="cIBXUqBHHKH0s7cT5Dt34zcMWJ0=">AAACD3icbVC7SgNBFJ2Nrxhfq5Y2g4magITdIGgZsLGSCOYB2SXMTu4mQ2YfzMwKYckf2PgrNhaK2Nra+TfOJilM9MDAmXPu5d57vJgzqSzr28itrK6tb+Q3C1vbO7t75v5BS0aJoNCkEY9ExyMSOAuhqZji0IkFkMDj0PZG15nffgAhWRTeq3EMbkAGIfMZJUpLPfO05MRDVnYCooaU8PR2UobK+cL3rFIp9cyiVbWmwH+JPSdFNEejZ345/YgmAYSKciJl17Zi5aZEKEY5TApOIiEmdEQG0NU0JAFIN53eM8EnWuljPxL6hQpP1d8dKQmkHAeerswWlcteJv7ndRPlX7kpC+NEQUhng/yEYxXhLBzcZwKo4mNNCBVM74rpkAhClY6woEOwl0/+S1q1qm1V7btasX4xjyOPjtAxKiMbXaI6ukEN1EQUPaJn9IrejCfjxXg3PmalOWPec4gWYHz+AJOVmwQ=</latexit>
<latexit sha1_base64="cIBXUqBHHKH0s7cT5Dt34zcMWJ0=">AAACD3icbVC7SgNBFJ2Nrxhfq5Y2g4magITdIGgZsLGSCOYB2SXMTu4mQ2YfzMwKYckf2PgrNhaK2Nra+TfOJilM9MDAmXPu5d57vJgzqSzr28itrK6tb+Q3C1vbO7t75v5BS0aJoNCkEY9ExyMSOAuhqZji0IkFkMDj0PZG15nffgAhWRTeq3EMbkAGIfMZJUpLPfO05MRDVnYCooaU8PR2UobK+cL3rFIp9cyiVbWmwH+JPSdFNEejZ345/YgmAYSKciJl17Zi5aZEKEY5TApOIiEmdEQG0NU0JAFIN53eM8EnWuljPxL6hQpP1d8dKQmkHAeerswWlcteJv7ndRPlX7kpC+NEQUhng/yEYxXhLBzcZwKo4mNNCBVM74rpkAhClY6woEOwl0/+S1q1qm1V7btasX4xjyOPjtAxKiMbXaI6ukEN1EQUPaJn9IrejCfjxXg3PmalOWPec4gWYHz+AJOVmwQ=</latexit>
<latexit sha1_base64="cIBXUqBHHKH0s7cT5Dt34zcMWJ0=">AAACD3icbVC7SgNBFJ2Nrxhfq5Y2g4magITdIGgZsLGSCOYB2SXMTu4mQ2YfzMwKYckf2PgrNhaK2Nra+TfOJilM9MDAmXPu5d57vJgzqSzr28itrK6tb+Q3C1vbO7t75v5BS0aJoNCkEY9ExyMSOAuhqZji0IkFkMDj0PZG15nffgAhWRTeq3EMbkAGIfMZJUpLPfO05MRDVnYCooaU8PR2UobK+cL3rFIp9cyiVbWmwH+JPSdFNEejZ345/YgmAYSKciJl17Zi5aZEKEY5TApOIiEmdEQG0NU0JAFIN53eM8EnWuljPxL6hQpP1d8dKQmkHAeerswWlcteJv7ndRPlX7kpC+NEQUhng/yEYxXhLBzcZwKo4mNNCBVM74rpkAhClY6woEOwl0/+S1q1qm1V7btasX4xjyOPjtAxKiMbXaI6ukEN1EQUPaJn9IrejCfjxXg3PmalOWPec4gWYHz+AJOVmwQ=</latexit>
<latexit sha1_base64="cIBXUqBHHKH0s7cT5Dt34zcMWJ0=">AAACD3icbVC7SgNBFJ2Nrxhfq5Y2g4magITdIGgZsLGSCOYB2SXMTu4mQ2YfzMwKYckf2PgrNhaK2Nra+TfOJilM9MDAmXPu5d57vJgzqSzr28itrK6tb+Q3C1vbO7t75v5BS0aJoNCkEY9ExyMSOAuhqZji0IkFkMDj0PZG15nffgAhWRTeq3EMbkAGIfMZJUpLPfO05MRDVnYCooaU8PR2UobK+cL3rFIp9cyiVbWmwH+JPSdFNEejZ345/YgmAYSKciJl17Zi5aZEKEY5TApOIiEmdEQG0NU0JAFIN53eM8EnWuljPxL6hQpP1d8dKQmkHAeerswWlcteJv7ndRPlX7kpC+NEQUhng/yEYxXhLBzcZwKo4mNNCBVM74rpkAhClY6woEOwl0/+S1q1qm1V7btasX4xjyOPjtAxKiMbXaI6ukEN1EQUPaJn9IrejCfjxXg3PmalOWPec4gWYHz+AJOVmwQ=</latexit>
K
r
(S
n
)
<latexit sha1_base64="1j+vTKWwEOjtLUwsPBcxMkWrZ8s=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvyk4pplq2ZNAReJnZMyyNF0zc/+IMRxQLjCDEnZs61IOQkSimJG0lI/liRCeIyGpKcpRwGRTjK9IoXHWhlAPxT6cQWn6u+OBAVSTgJPV2ZbynkvE//zerHyz52E8ihWhOPZID9mUIUwiwQOqCBYsYkmCAuqd4V4hATCSgdX0iHY8ycvkna9Zls1+7pebpzmcRTBATgCVWCDM9AAl6AJWgCDB/AEXsCr8Wg8G2/G+6y0YOQ9++APjI9v1IaYEw==</latexit>
<latexit sha1_base64="1j+vTKWwEOjtLUwsPBcxMkWrZ8s=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvyk4pplq2ZNAReJnZMyyNF0zc/+IMRxQLjCDEnZs61IOQkSimJG0lI/liRCeIyGpKcpRwGRTjK9IoXHWhlAPxT6cQWn6u+OBAVSTgJPV2ZbynkvE//zerHyz52E8ihWhOPZID9mUIUwiwQOqCBYsYkmCAuqd4V4hATCSgdX0iHY8ycvkna9Zls1+7pebpzmcRTBATgCVWCDM9AAl6AJWgCDB/AEXsCr8Wg8G2/G+6y0YOQ9++APjI9v1IaYEw==</latexit>
<latexit sha1_base64="1j+vTKWwEOjtLUwsPBcxMkWrZ8s=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvyk4pplq2ZNAReJnZMyyNF0zc/+IMRxQLjCDEnZs61IOQkSimJG0lI/liRCeIyGpKcpRwGRTjK9IoXHWhlAPxT6cQWn6u+OBAVSTgJPV2ZbynkvE//zerHyz52E8ihWhOPZID9mUIUwiwQOqCBYsYkmCAuqd4V4hATCSgdX0iHY8ycvkna9Zls1+7pebpzmcRTBATgCVWCDM9AAl6AJWgCDB/AEXsCr8Wg8G2/G+6y0YOQ9++APjI9v1IaYEw==</latexit>
<latexit sha1_base64="1j+vTKWwEOjtLUwsPBcxMkWrZ8s=">AAACBXicbVDLSsNAFJ3UV62vqEtdDLZC3ZSkCLosuBHcVLQPaGOYTCft0MkkzEyEErJx46+4caGIW//BnX/jpI2grQcGzpxzL/fe40WMSmVZX0ZhaXllda24XtrY3NreMXf32jKMBSYtHLJQdD0kCaOctBRVjHQjQVDgMdLxxheZ37knQtKQ36pJRJwADTn1KUZKS655WOkHSI08P7lK70T153OTuvyk4pplq2ZNAReJnZMyyNF0zc/+IMRxQLjCDEnZs61IOQkSimJG0lI/liRCeIyGpKcpRwGRTjK9IoXHWhlAPxT6cQWn6u+OBAVSTgJPV2ZbynkvE//zerHyz52E8ihWhOPZID9mUIUwiwQOqCBYsYkmCAuqd4V4hATCSgdX0iHY8ycvkna9Zls1+7pebpzmcRTBATgCVWCDM9AAl6AJWgCDB/AEXsCr8Wg8G2/G+6y0YOQ9++APjI9v1IaYEw==</latexit>
Figure 3: The neighbor-view interactions. We compute a similarity
matrix between the neighbors N (e) and N (e
′
) and apply a dual
aggregation function to extract the features form the matrix.
is to a given neighbor of e. In another word, a neighbor of e is
not required to be similar to all the neighbors of e
′
, thus other
similarities except the maximal one are discarded. Next, the
maximal value s
max
i
is transformed into a row-based feature
vector K
r
(S
i
), with each of the l-th element K
l
(s
max
i
) being
converted by the l-th RBF kernel with mean µ
l
and variance
σ
l
. Then K
r
(S
i
) of all the rows are averaged into a row-based
similarity embedding ϕ
r
(N (e), N (e
′
)):
s
max
i
=
n
max
j=0
{s
i0
, · · · , s
ij
, · · · , s
in
} (3)
K
l
(s
max
i
) = exp
[
−
(s
max
i
− µ
l
)
2
2σ
2
l
]
K
r
(S
i
) = [ K
1
(s
max
i
), · · · , K
l
(s
max
i
), · · · , K
L
(s
max
i
)]
ϕ
r
(N (e), N (e
′
)) =
1
|N (e)|
|N (e)|
∑
i=1
log K
r
(S
i
),
where n is the maximal number of neighbors over all the en-
tities. S will be padded zero if the real number |N (e)| is less
than n. The kernels are used to convert the one-dimension
similarity into L-dimension similarities to enhance the dis-
crimination ability, where the kernel with µ = 1 and σ → 0
only considers the exact matches between neighbors, and oth-
ers capture the semantic matches between neighbors.
The above process aggregates the features along the rows,
which reflects how similar each neighbor e
i
∈ N (e) is to the
neighbors N (e
′
) of e
′
. Similarly, we also aggregate the fea-
tures along the columns to capture how similar each neighbor
e
′
j
∈ N (e
′
) is to the neighbors N (e) of e. Finally, the row-
aggregation vector ϕ
r
and the column-aggregation vector ϕ
c
are concatenated as the final similarity embedding:
ϕ(N (e), N (e
′
))=ϕ
r
(N (e), N (e
′
)) ⊕ ϕ
c
(N (e), N (e
′
)), (4)
where ⊕ indicates the concatenation operation. Figure 3 il-
lustrates the neighbor-view interactions.
Neighboring Relation Mask Matrix
A pair of neighbors are more convincing for supporting the
alignment of two entities, if the corresponding relations are
Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence (IJCAI-20)
3176
评论0