没有合适的资源？快使用搜索试试~ 我知道了~

文库首页人工智能深度学习2021.7.20_NeurIPS-2020-introducing-routing-uncertainty

2021.7.20_NeurIPS-2020-introducing-routing-uncertainty

routing

需积分: 0 0 下载量 90 浏览量 2021-07-26 20:40:18 上传评论收藏 4.89MB PPTX 举报

温馨提示

试读

36页

2021.7.20_NeurIPS-2020-introducing-routing-uncertainty

资源详情

资源评论

资源推荐

34t h C onf eren ce on Ne ural I nfo rma tio n P ro c es s in g S yst em s (Neu rIP S

2020 ), Vanc ouve r, Canad a.

Introduction

Weaknesses of Capsule Networks:

1. The memory bottleneck incurred by vector valued activations;

2. Iterative nature of capsule routing algorithms results in inefficient models;

3. They are also prone to underfitting or overfitting if the number of routing iterations

isn’t properly set

To address the above weaknesses one may decide to naively replace the iterative nature

of capsule routing with some faster alternative.

Introduction-motivations

4 core premises of Capsule networks:

1. Whether viewpoint-invariance and affine transformation robustness

properties are retained;

2. Changes in assumptions about part-object relationships are made explicit;

3. Whether capsules are still activated based on high-dimensional coincidences;

4. How do we handle the intrinsic uncertainty in assembling parts into objects.

we focus on the core premises of capsule networks, and on enhancing their

advantages over CNNs: viewpoint-invariance, and affine transformation robustness

whilst being more efficient.

Introduction-Contribution

Rather than performing local iterative routing between adjacent capsule layers which is

inefficient, we propose an alternative global view based on representing the inherent

uncertainty in part-object relationships, by approximating a posterior distribution over

part-object connections.

Sources of uncertainty in assembling objects via a composition of parts can arise from

numerous sources, such as:

1. feature occlusions due to observed viewpoints;

2. sensory noise in captured data;

3. object symmetries for which poses may be ambiguous such as spherical objects/parts.

剩余35页未读，继续阅读

Introducing_Data_Science_Big_Data-_Machine_Learning-_and_more-_using_Python_tool

Introducing_Data_Science_Big_Data-_Machine_Learning-_and_more-_using_Python_tools

Flink-Forward-SF-2017_Eron-Wright_Introducing-Flink-TensorFlow

Slide share: Flink-Forward-SF-2017_Eron-Wright_Introducing-Flink-TensorFlow

Forrester-2021_01-Introducing-The-Zero-Trust-Edge-Model

Forrester-2021_01-Introducing-The-Zero-Trust-Edge-Model

en_microsoft_press_e-book_introducing_windows_server_2008_chapter_1_introduction

en_microsoft_press_e-book_introducing_windows_server_2008_chapter_1_introduction

内容反馈

NamoAmitabhaOrz

粉丝: 0
资源: 1

上传资源快速赚钱

我的内容管理展开

我的资源快来上传第一个资源

我的收益

登录查看自己的收益

我的积分登录查看自己的积分

我的C币登录后查看C币余额

我的收藏

我的下载

下载帮助

前往需求广场，查看用户热搜

2021.7.20_NeurIPS-2020-introducing-routing-uncertainty

评论0

最新资源

2021.7.20_NeurIPS-2020-introducing-routing-uncertainty

评论0

Introducing_Data_Science_Big_Data-_Machine_Learning-_and_more-_using_Python_tool

Flink-Forward-SF-2017_Eron-Wright_Introducing-Flink-TensorFlow

Forrester-2021_01-Introducing-The-Zero-Trust-Edge-Model

en_microsoft_press_e-book_introducing_windows_server_2008_chapter_1_introduction

learn_the_architecture_-_introducing_amba_chi_102407_0100_01_en.pdf

LTM Config 配置手册打包

Introducing_Low-Density_Parity-Check_Codes

ci-plus_specification_v1-4-3.pdf

Python_End-to-end Data Analysis-Packt Publishing(2017).pdf

Hello Android.rar

CSharp 3.0 With the .NET Framework 3.5 Unleashed（english）

Introducing_The_Pentaho_Workbench_zh_CN_1.0.0.0.pdf

Linux的英文原版书（Ubuntu）part1

ASEF-Blog_ASEF-Blog(4).pdf

nokia Introducing_Web_Run-Time_v1_1_zh_CH

GPS.rar_GPS_GPS c++_GPS.C_GPX..cpp_it

Linux的英文原版书（Ubuntu） part2

Multimodal Deep Learning

AngularJS - Novice to Ninja.pdf.pdf )

YOLOv8-deepsort 实现智能车辆目标检测+车辆跟踪+车辆计数

YOLOv8网络结构图，自制visio文件，yolov8.vsds，需要的自取，在原有的基础上直接改就行了

yolov8(2023年8月版本),已经下好yolov8s.pt和yolov8n.pt

Transformer模型实现长期预测并可视化结果（附代码+数据集+原理介绍）

社交平台上经济类话题的文章热度信息，数据是真实的，但不是真实日期

行人跌倒数据集（VOC格式）

Unet眼底血管图像分割数据集+代码+模型+系统界面+教学视频.zip

YOLOV5 + 双目相机实现三维测距（新版本）

全新的SOTA模型YOLOv9

YOLOV5口罩检测数据集+代码+模型 2000张标注好的数据+教学视频.zip

最新资源