最前沿的深度學習論文、架構及資源分享

    深度摺積神經網路在影象、語音及NLP領域取得了巨大的成功，從學習和分享的角度出發，本篇文章整理了自2013年以來關於CNN相關的最新的資源，包括重要的論文、書籍、影片教程、Tutorial、理論、模型庫和開發庫。文末附連結版資源地址。

重要的論文：

    1. Very deep convolutional networks for large-scale image recognition (VGG-net) (2014)

    2. Going deeper with convolutions (GoogLeNet) by Google (2015)

    3. Deep learning (2015)

    4. Visualizing and Understanding Convolutional Neural Networks (ZF Net) (2014)

    5. Fully convolutional networks for semantic segmentation (2015)

    6. Deep residual learning for image recognition (ResNet) by Microsoft (2015)

    7. Deepface closing the gap to human-level performance in face verification (2014)

    8. Batch normalization Accelerating deep network training by reducing internal covariate shift (2015)

    9. Deep Learning in Neural Networks An Overview (2015)

    10. Delving deep into rectifiers Surpassing human-level performance on imagenet classification (PReLU) (2014)

    11. Faster R-CNN Towards real-time object detection with region proposal networks (2015)

    12. Fast R-CNN (2015)

    13. Spatial pyramid pooling in deep convolutional networks for visual recognition (SPP Net) (2014)

    14. Generative Adversarial Nets (2014)

    15. Spatial Transformer Networks (2015)

    16. Understanding deep image representations by inverting them (2015)

    17. Deep Learning of Representations Looking Forward (2013)

經典的文章：

    18. mageNet Classification with Deep Convolutional Neural Networks (AlexNet) (2012)

    19. Rectified linear units improve restricted boltzmann machines (ReLU) (2010)

重要的理論：

    20. Deep Neural Networks are Easily Fooled High Confidence Predictions for Unrecognizable Images (2015)

    21. Distilling the Knowledge in a Neural Network (2015)

    22. Deep learning in neural networks An overview (2015)

重要的書籍：

    23. Deep Learning Textbook – An MIT Press book (2016)

    24. Learning Deep Architectures for AI

    25. Neural Nets and Deep Learning

重要的課程/Tutorial：

    26. Caffe Tutorial (CVPR 2015)

    27. Tutorial on Deep Learning for Vision (CVPR 2014)

    28. Introduction to Deep Learning with Python – Theano Tutorials

    29. Deep Learning Tutorials with Theano/Python

    30. Deep Learning Take machine learning to the next level (by udacity)

    31. DeepLearnToolbox – A Matlab toolbox for Deep Learning

    32. Stanford Matlab-based Deep Learning

    33. Stanford 231n Class Convolutional Neural Networks for Visual Recognition

    34. Deep Learning Course (by Yann LeCun-2016)

    35. Generative Models (by OpenAI)

    36. An introduction to Generative Adversarial Networks (with code in TensorFlow)

重要的資源/模型：

    37. VGG-net

    38. GoogLeNet

    39. ResNet – MatConvNet implementation

    40. AlexNet

    41. Fully Convolutional Networks for Semantic Segmentation

    42. OverFeat

    43. SPP_net

    44. Fast R-CNN

    45. Faster R-CNN

    46. Generative Adversarial Networks (GANs)

    47. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks)

    48. ResNeXt Aggregated Residual Transformations for Deep Neural Networks)

    49. MultiPath Network training code

重要的架構和開發庫：

    50. Tensorflow by Google [C++ and CUDA]

    51. Caffe by Berkeley Vision and Learning Center (BVLC) [C++][Installation Instructions]

    52. Keras by François Chollet [Python]

    53. Microsoft Cognitive Toolkit – CNTK [C++]

    54. MXNet adapted by Amazon [C++]

    55. Torch by Collobert, Kavukcuoglu & Clement Farabet, widely used by Facebook [Lua]

    56. Convnetjs by Andrej Karpathy [JavaScript]

    57. Theano by Université de Montréal [Python]

    58. Deeplearning4j by startup Skymind [Java]

    59. Paddle by Baidu [C++]

    60. Deep Scalable Sparse Tensor Network Engine (DSSTNE) by Amazon [C++]

    61. Neon by Nervana Systems [Python & Sass]

    62. Chainer [Python]

    63. h2o [Java]

    64. Brainstorm by Istituto Dalle Molle di Studi sull’Intelligenza Artificiale (IDSIA) [Python]

    65. Matconvnet by Andrea Vedaldi [Matlab]