DNN神经网络讲解

所需积分/C币:48 2018-07-14 20:32:59 2.41MB PDF

李宏毅教授关于对DNN深度神经网络的讲解,适合深度学习初学者学习使用。
Do not always blame overfitting 20 Not well trained 20 56-layer a50 20-layer 56-layer Over (Ing 20-layer 0 0 6 6 Iter (Ie4 iter. (le4) Training Data Testing Data Deep residual learning for image recognition http://arxiv.org/abs/1512.03385 Recipe of Deep Learning YES Good results on Different approaches for Testing Data? different problems YES e.g. dropout for good results on testing data Good results on Training data? Neural Network Recipe of Deep Learning YES Early Stopping Good results on Regularization Testing Data? Dropout YES Good results on New activation function Training data? Adaptive Learning Rate Hard to get the power of deep Handwritting Digit Classification 100 5 85 80 75 Results on Training data 60 Deeper usually does not imply better 2 3 4 5 6 7 8 9 Layers Vanishing gradient problem y1 Smaller gradients Larger gradients Learn very slow Learn very fast Almost random Already converge based on random!? Vanishing gradient problem Smaller gradients Small ■■日■■■■■■■口■■■■■■ output 1 arge +△W npu Intuitive way to compute the derivatives 0C△C 0w△w Relu Rectified Linear Unit(ReLU) Reason C F(Z 1. Fast to compute C三z 2. Biological reason a=0 3. Infinite sigmoid with different biases 4. Vanishing gradient Xavier Glorot, AlSTATS'11 [Andrew L Maas ICML 13 problem [ Kaiming he,arXⅳv15] C C三z Relu a=0 0 0

...展开详情
试读 49P DNN神经网络讲解
img
涂涂凉凉

关注 私信 TA的资源

上传资源赚积分,得勋章
    最新推荐
    DNN神经网络讲解 48积分/C币 立即下载
    1/49
    DNN神经网络讲解第1页
    DNN神经网络讲解第2页
    DNN神经网络讲解第3页
    DNN神经网络讲解第4页
    DNN神经网络讲解第5页
    DNN神经网络讲解第6页
    DNN神经网络讲解第7页
    DNN神经网络讲解第8页
    DNN神经网络讲解第9页
    DNN神经网络讲解第10页
    DNN神经网络讲解第11页
    DNN神经网络讲解第12页
    DNN神经网络讲解第13页
    DNN神经网络讲解第14页
    DNN神经网络讲解第15页

    试读已结束,剩余34页未读...

    48积分/C币 立即下载 >