搜索资源列表
Learning-Deep-Architectures
- 深度学习经典教程,PPT格式,强烈推荐,值得一读-Deep learning of classic tutorial, PPT format, it is strongly recommended
deep-learning-study-by-xiaoda
- 浅谈深度学习,肖达经典讲义,深入浅出,值得学习-On the depth of learning, Xiaoda classic handouts, easy to understand, it is worth learning
Introduction-to-13deep-learning
- 深度学习是机器学习最热门的领域,适合机器视觉的深度学习-ntroduction to deep learning notes for machine vision, deep learning is now one of the most popular field of machine learning field
Deep-Learningdoc
- 一篇关于对深度网络工具箱源代码调试和理解的文档,适合初学深度学习理论的研究生-a document on the coding and running matlab soucecode on deep learning, it is very good for the new learner of deep learn
matlab-Deep-Learn
- MATLAB神经网络经典程序 包括多个神经网络程序、深度学习程序-MATLAB neural network program including mutil network and deep learning program
Neural-Networks-and-Deep-Learning
- 很好的深度学习与神经网络教程,适合深度学习的初学者-Good depth and neural network learning tutorial for beginners to learn the depth of
deep-residual-networks-master
- 深度残差网络的介绍与源代码,适合深度学习爱好者学习。这是何凯明大牛的又一部大作。-The depth of the residual network is introduced with the source code , suitable for deep learning lovers to study . This is another masterpiece He Kaiming Daniel .
Deep-Learning-Tutorial
- 深度学习最入门的文档,一天可以搞懂,李宏毅著-Depth learning the most entry-level documents, one day can get to know, Li Hongyi with
DeepNeuralNetwork20150129
- deep neural learning
Overview of Multi Task Learning
- 这是一篇综述,主要是关于基于深度学习的多任务学习的一些进展。(An Overview of Multi-Task Learning in Deep Neural Networks)
DeeBNetV3.2
- matlab's deep belief network toolbox
1509.02971
- pdf file about deep reinforcement learning
2016-MLSS-RL
- pdf file about deep reinforcement learning
GoogleNet_MATLAB-master
- GoogleNet 卷积神经网络 图片分类 分类精度高 网络结构深(GoogleNet convolution neural network image classification, high classification accuracy, network structure is deep)
Deep Learning Book
- 深度学习官方离线文档,专业学习资料,人工智能进阶学术文档。(Artificial intelligence official offline documents, professional learning materials)
CNTK
- 在深度的重要性的驱使下,出现了一个新的问题:训练一个更好的网络是否和堆叠更多的层一样简单呢?解决这一问题的障碍便是困扰人们很久的梯度消失/梯度爆炸,这从一开始便阻碍了模型的收敛。归一初始化(normalized initialization)和中间归一化(intermediate normalization)在很大程度上解决了这一问题,它使得数十层的网络在反向传播的随机梯度下降(SGD)上能够收敛。 当深层网络能够收敛时,一个退化问题又出现了:随着网络深度的增加,准确率达到饱和(不足为奇)然后迅
deeplearningbook-chinese-0.5-beta
- 卷积神经网络、循环神经网络、递归神经网络、深度信念网络、深度堆叠网络、LSTM长短时记忆(Convolution neural network, circulating neural network, recursive neural network, deep belief network, deep stack network, LSTM length memory)
DeepLearnToolbox-master
- 深度学习工具箱,包括卷积神经网络,DBN等,可以运行tests文件碱性测试所有的代码(Depth learning toolbox)
lec1
- Machine Learning Basics
rl03
- Lecture 3: Q-learning (table)