126篇殿堂級深度學習論文分類整理，從入門到應用

來源：雷課

如果你有非常大的決心從事深度學習，又不想在這一行打醬油，那麼研讀大牛論文將是不可避免的一步。而作為新人，你的第一個問題或許是：“論文那麼多，從哪一篇讀起？”本文將試圖解決這個問題——文章標題本來是：“從入門到絕望，無止境的深度學習論文”。請諸位備好道具，開啟頭懸梁錐刺股的學霸姿勢。

開個玩笑。但對非科班出身的開發者而言，讀論文的確可以成為一件很痛苦的事。但好訊息來了——為避免初學者陷入迷途苦海，暱稱為 songrotek 的學霸在 GitHub 釋出了他整理的深度學習路線圖，分門別類梳理了新入門者最需要學習的 DL 論文，又按重要程度給每篇論文打上星星。

截至目前，這份 DL 論文路線圖已在 GitHub 收穫了近萬顆星星好評，人氣極高。小雷感到非常有必要對大家進行介紹。閒話少說，該路線圖根據以下四項原則而組織：從大綱到細節、從經典到前沿、從一般到具體領域、關注最新研究突破。有許多論文很新但非常值得一讀。

下文內容連結比較多，建議收藏起來慢慢看

1 深度學習歷史和基礎

1.0 書籍

[0] Bengio, Yoshua, Ian J. Goodfellow, and Aaron Courville."Deep learning." An MIT Press book. (2015). [pdf] (Ian Goodfellow 等大牛所著的教科書，乃深度學習聖經。你可以同時研習這本書以及以下論文)

地址：
https://github.com/HFTrader/DeepLearningBook/raw/master/DeepLearningBook.pdf

1.1 調查

[1] LeCun, Yann, Yoshua Bengio, and Geoffrey Hinton. "Deep learning." Nature 521.7553 (2015): 436-444. [pdf] (三巨頭做的調查)

地址：
http://www.cs.toronto.edu/~hinton/absps/NatureDeepReview.pdf

1.2 深度置信網路 (DBN，深度學習前夜的里程碑)

[2] Hinton, Geoffrey E., Simon Osindero, and Yee-Whye Teh. "A fast learning algorithm for deep belief nets." Neural computation 18.7 (2006): 1527-1554. [pdf] (深度學習前夜)

地址：
http://www.cs.toronto.edu/~hinton/absps/ncfast.pdf

[3] Hinton, Geoffrey E., and Ruslan R. Salakhutdinov. "Reducing the dimensionality of data with neural networks." Science 313.5786 (2006): 504-507. [pdf] (里程碑，展示了深度學習的前景)

地址：
http://www.cs.toronto.edu/~hinton/science.pdf

1.3 ImageNet 的進化（深度學習從此萌發）

[4] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification with deep convolutional neural networks." Advances in neural information processing systems. 2012. [pdf] (AlexNet, 深度學習突破)

地址：
http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf

[5] Simonyan, Karen, and Andrew Zisserman. "Very deep convolutional networks for large-scale image recognition." arXiv preprint arXiv:1409.1556 (2014). [pdf] (VGGNet，神經網路變得很深層)

地址：
https://arxiv.org/pdf/1409.1556.pdf

[6] Szegedy, Christian, et al. "Going deeper with convolutions." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015. [pdf] (GoogLeNet)

地址：
http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Szegedy_Going_Deeper_With_2015_CVPR_paper.pdf

[7] He, Kaiming, et al. "Deep residual learning for image recognition." arXiv preprint arXiv:1512.03385 (2015). [pdf](ResNet，特別深的神經網路, CVPR 最佳論文)

地址：
https://arxiv.org/pdf/1512.03385.pdf

1.4 語音識別的進化

[8] Hinton, Geoffrey, et al. "Deep neural networks for acoustic modeling in speech recognition: The shared views of four research groups." IEEE Signal Processing Magazine 29.6 (2012): 82-97. [pdf] (語音識別的突破)

地址：

http://cs224d.stanford.edu/papers/maas_paper.pdf

[9] Graves, Alex, Abdel-rahman Mohamed, and Geoffrey Hinton. "Speech recognition with deep recurrent neural networks." 2013 IEEE international conference on acoustics, speech and signal processing. IEEE, 2013. [pdf] (RNN)

地址：

http://arxiv.org/pdf/1303.5778.pdf

[10] Graves, Alex, and Navdeep Jaitly. "Towards End-To-End Speech Recognition with Recurrent Neural Networks." ICML. Vol. 14. 2014. [pdf]

地址：
http://www.jmlr.org/proceedings/papers/v32/graves14.pdf

[11] Sak, Haim, et al. "Fast and accurate recurrent neural network acoustic models for speech recognition." arXiv preprint arXiv:1507.06947 (2015). [pdf] (谷歌語音識別系統)

地址：
http://arxiv.org/pdf/1507.06947

[12] Amodei, Dario, et al. "Deep speech 2: End-to-end speech recognition in english and mandarin." arXiv preprint arXiv:1512.02595 (2015). [pdf] (百度語音識別系統)

地址：
https://arxiv.org/pdf/1512.02595.pdf

[13] W. Xiong, J. Droppo, X. Huang, F. Seide, M. Seltzer, A. Stolcke, D. Yu, G. Zweig "Achieving Human Parity in Conversational Speech Recognition." arXiv preprint arXiv:1610.05256 (2016). [pdf] (最前沿的語音識別, 微軟)

地址：
https://arxiv.org/pdf/1610.05256v1

研讀以上論文之後，你將對深度學習歷史、模型的基本架構（包括 CNN, RNN, LSTM）有一個基礎的瞭解，並理解深度學習如何應用於影像和語音識別問題。接下來的論文，將帶你深入探索深度學習方法、在不同領域的應用和前沿尖端技術。我建議，你可以根據興趣和工作/研究方向進行選擇性的閱讀。

2 深度學習方法

2.1 模型

[14] Hinton, Geoffrey E., et al. "Improving neural networks by preventing co-adaptation of feature detectors." arXiv preprint arXiv:1207.0580 (2012). [pdf] (Dropout)

地址：
https://arxiv.org/pdf/1207.0580.pdf

[15] Srivastava, Nitish, et al. "Dropout: a simple way to prevent neural networks from overfitting." Journal of Machine Learning Research 15.1 (2014): 1929-1958. [pdf]

地址：
http://www.jmlr.org/papers/volume15/srivastava14a.old/source/srivastava14a.pdf

[16] Ioffe, Sergey, and Christian Szegedy. "Batch normalization: Accelerating deep network training by reducing internal covariate shift." arXiv preprint arXiv:1502.03167 (2015). [pdf] (2015 年的傑出研究)

地址：
http://arxiv.org/pdf/1502.03167

[17] Ba, Jimmy Lei, Jamie Ryan Kiros, and Geoffrey E. Hinton. "Layer normalization." arXiv preprint arXiv:1607.06450 (2016). [pdf] (Batch Normalization 的更新)

地址：
https://arxiv.org/pdf/1607.06450.pdf?utm_source=sciontist.com&utm_medium=refer&utm_campaign=promote

[18] Courbariaux, Matthieu, et al. "Binarized Neural Networks: Training Neural Networks with Weights and Activations Constrained to+ 1 or1." [pdf] (新模型，快)

地址：
https://pdfs.semanticscholar.org/f832/b16cb367802609d91d400085eb87d630212a.pdf

[19] Jaderberg, Max, et al. "Decoupled neural interfaces using synthetic gradients." arXiv preprint arXiv:1608.05343 (2016). [pdf] (訓練方法的創新，研究相當不錯)

地址：
https://arxiv.org/pdf/1608.05343

[20] Chen, Tianqi, Ian Goodfellow, and Jonathon Shlens. "Net2net: Accelerating learning via knowledge transfer." arXiv preprint arXiv:1511.05641 (2015). [pdf] (改進此前的訓練網路，來縮短訓練週期)

地址：
https://arxiv.org/abs/1511.05641

[21] Wei, Tao, et al. "Network Morphism." arXiv preprint arXiv:1603.01670 (2016). [pdf] (改進此前的訓練網路，來縮短訓練週期)

地址：
https://arxiv.org/abs/1603.01670

2.2 優化 Optimization

[22] Sutskever, Ilya, et al. "On the importance of initialization and momentum in deep learning." ICML (3) 28 (2013): 1139-1147. [pdf] (Momentum optimizer)

地址：
http://www.jmlr.org/proceedings/papers/v28/sutskever13.pdf

[23] Kingma, Diederik, and Jimmy Ba. "Adam: A method for stochastic optimization." arXiv preprint arXiv:1412.6980 (2014). [pdf] (Maybe used most often currently)

地址：
http://arxiv.org/pdf/1412.6980

[24] Andrychowicz, Marcin, et al. "Learning to learn by gradient descent by gradient descent."arXiv preprint arXiv:1606.04474 (2016). [pdf] (Neural Optimizer,Amazing Work)

地址：
https://arxiv.org/pdf/1606.04474

[25] Han, Song, Huizi Mao, and William J. Dally."Deep compression: Compressing deep neural network with pruning, trained quantization and huffman coding." CoRR, abs/1510.00149 2 (2015). [pdf] (ICLR best paper, new direction to make NN running fast,DeePhi Tech Startup)

地址：
https://pdfs.semanticscholar.org/5b6c/9dda1d88095fa4aac1507348e498a1f2e863.pdf

[26] Iandola, Forrest N., et al. "SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 1MB model size." arXiv preprint arXiv:1602.07360 (2016). [pdf] (Also a new direction to optimize NN,DeePhi Tech Startup)

地址：
http://arxiv.org/pdf/1602.07360

2.3 無監督學習/深度生成模型

[27] Le, Quoc V. "Building high-level features using large scale unsupervised learning." 2013 IEEE international conference on acoustics, speech and signal processing. IEEE, 2013. [pdf] (里程碑, 吳恩達, 谷歌大腦, Cat)

地址：
http://arxiv.org/pdf/1112.6209.pdf&embed

[28] Kingma, Diederik P., and Max Welling. "Auto-encoding variational bayes." arXiv preprint arXiv:1312.6114 (2013). [pdf](VAE)

地址：
http://arxiv.org/pdf/1312.6114

[29] Goodfellow, Ian, et al. "Generative adversarial nets." Advances in Neural Information Processing Systems. 2014. [pdf](GAN，很酷的想法)

地址：
http://papers.nips.cc/paper/5423-generative-adversarial-nets.pdf

[30] Radford, Alec, Luke Metz, and Soumith Chintala. "Unsupervised representation learning with deep convolutional generative adversarial networks." arXiv preprint arXiv:1511.06434 (2015). [pdf] (DCGAN)

地址：
http://arxiv.org/pdf/1511.06434

[31] Gregor, Karol, et al. "DRAW: A recurrent neural network for image generation." arXiv preprint arXiv:1502.04623 (2015). [pdf] (VAE with attention, 很出色的研究)

地址：
http://jmlr.org/proceedings/papers/v37/gregor15.pdf

[32] Oord, Aaron van den, Nal Kalchbrenner, and Koray Kavukcuoglu. "Pixel recurrent neural networks." arXiv preprint arXiv:1601.06759 (2016). [pdf] (PixelRNN)

地址：
http://arxiv.org/pdf/1601.06759

[33] Oord, Aaron van den, et al."Conditional image generation with PixelCNN decoders." arXiv preprint arXiv:1606.05328 (2016). [pdf] (PixelCNN)

地址：
https://arxiv.org/pdf/1606.05328

2.4 遞迴神經網路（RNN） / Sequence-to-Sequence Model

[34] Graves, Alex. "Generating sequences with recurrent neural networks." arXiv preprint arXiv:1308.0850 (2013). [pdf] (LSTM, 效果很好，展示了 RNN 的效能)

地址：
http://arxiv.org/pdf/1308.0850

[35] Cho, Kyunghyun, et al. "Learning phrase representations using RNN encoder-decoder for statistical machine translation." arXiv preprint arXiv:1406.1078 (2014). [pdf] (第一篇 Sequence-to-Sequence 的論文)

地址：
http://arxiv.org/pdf/1406.1078

[36] Sutskever, Ilya, Oriol Vinyals, and Quoc V. Le. "Sequence to sequence learning with neural networks." Advances in neural information processing systems. 2014. [pdf] (傑出研究)

地址：
http://papers.nips.cc/paper/5346-information-based-learning-by-agents-in-unbounded-state-spaces.pdf

[37] Bahdanau, Dzmitry, KyungHyun Cho, and Yoshua Bengio. "Neural Machine Translation by Jointly Learning to Align and Translate." arXiv preprint arXiv:1409.0473 (2014). [pdf]

地址：
https://arxiv.org/pdf/1409.0473v7.pdf

[38] Vinyals, Oriol, and Quoc Le. "A neural conversational model." arXiv preprint arXiv:1506.05869 (2015). [pdf] (Seq-to-Seq 聊天機器人)

地址：
http://arxiv.org/pdf/1506.05869.pdf%20(http://arxiv.org/pdf/1506.05869.pdf)

2.5 神經網路圖靈機

[39] Graves, Alex, Greg Wayne, and Ivo Danihelka. "Neural turing machines." arXiv preprint arXiv:1410.5401 (2014). [pdf] (未來計算機的基礎原型機)

地址：
http://arxiv.org/pdf/1410.5401.pdf

[40] Zaremba, Wojciech, and Ilya Sutskever. "Reinforcement learning neural Turing machines." arXiv preprint arXiv:1505.00521 362 (2015). [pdf]

地址：
https://pdfs.semanticscholar.org/f10e/071292d593fef939e6ef4a59baf0bb3a6c2b.pdf

[41] Weston, Jason, Sumit Chopra, and Antoine Bordes. "Memory networks." arXiv preprint arXiv:1410.3916 (2014). [pdf]

地址：
http://arxiv.org/pdf/1410.3916

[42] Sukhbaatar, Sainbayar, Jason Weston, and Rob Fergus. "End-to-end memory networks." Advances in neural information processing systems. 2015. [pdf]

地址：
http://papers.nips.cc/paper/5846-end-to-end-memory-networks.pdf

[43] Vinyals, Oriol, Meire Fortunato, and Navdeep Jaitly. "Pointer networks." Advances in Neural Information Processing Systems. 2015. [pdf]

地址：
http://papers.nips.cc/paper/5866-pointer-networks.pdf

[44] Graves, Alex, et al. "Hybrid computing using a neural network with dynamic external memory." Nature (2016). [pdf] (里程碑，把以上論文的想法整合了起來)

地址：
https://www.dropbox.com/s/0a40xi702grx3dq/2016-graves.pdf

2.6 深度強化學習

[45] Mnih, Volodymyr, et al. "Playing atari with deep reinforcement learning." arXiv preprint arXiv:1312.5602 (2013). [pdf]) (第一個以深度強化學習為題的論文)

地址：
http://arxiv.org/pdf/1312.5602.pdf

[46] Mnih, Volodymyr, et al. "Human-level control through deep reinforcement learning." Nature 518.7540 (2015): 529-533. [pdf] (里程碑)

地址：
https://storage.googleapis.com/deepmind-data/assets/papers/DeepMindNature14236Paper.pdf

[47] Wang, Ziyu, Nando de Freitas, and Marc Lanctot. "Dueling network architectures for deep reinforcement learning." arXiv preprint arXiv:1511.06581 (2015). [pdf] (ICLR 最佳論文，很棒的想法)

地址：
http://arxiv.org/pdf/1511.06581

[48] Mnih, Volodymyr, et al. "Asynchronous methods for deep reinforcement learning." arXiv preprint arXiv:1602.01783 (2016). [pdf] (前沿方法)

地址：
http://arxiv.org/pdf/1602.01783

[49] Lillicrap, Timothy P., et al. "Continuous control with deep reinforcement learning." arXiv preprint arXiv:1509.02971 (2015). [pdf] (DDPG)

地址：
http://arxiv.org/pdf/1509.02971

[50] Gu, Shixiang, et al. "Continuous Deep Q-Learning with Model-based Acceleration." arXiv preprint arXiv:1603.00748 (2016). [pdf] (NAF)

地址：
http://arxiv.org/pdf/1603.00748

[51] Schulman, John, et al. "Trust region policy optimization." CoRR, abs/1502.05477 (2015). [pdf] (TRPO)

地址：
http://www.jmlr.org/proceedings/papers/v37/schulman15.pdf

[52] Silver, David, et al. "Mastering the game of Go with deep neural networks and tree search." Nature 529.7587 (2016): 484-489. [pdf] (AlphaGo)

地址：
http://willamette.edu/~levenick/cs448/goNature.pdf

2.7 深度遷移學習 /終生學習 / 強化學習

[53] Bengio, Yoshua. "Deep Learning of Representations for Unsupervised and Transfer Learning." ICML Unsupervised and Transfer Learning 27 (2012): 17-36. [pdf] (這是一個教程)

地址：
http://www.jmlr.org/proceedings/papers/v27/bengio12a/bengio12a.pdf

[54] Silver, Daniel L., Qiang Yang, and Lianghao Li. "Lifelong Machine Learning Systems: Beyond Learning Algorithms." AAAI Spring Symposium: Lifelong Machine Learning. 2013. [pdf] (對終生學習的簡單討論)

地址：
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.696.7800&rep=rep1&type=pdf

[55] Hinton, Geoffrey, Oriol Vinyals, and Jeff Dean. "Distilling the knowledge in a neural network." arXiv preprint arXiv:1503.02531 (2015). [pdf] (大神們的研究)

地址：
http://arxiv.org/pdf/1503.02531

[56] Rusu, Andrei A., et al. "Policy distillation." arXiv preprint arXiv:1511.06295 (2015). [pdf] (RL 領域)

地址：
http://arxiv.org/pdf/1511.06295

[57] Parisotto, Emilio, Jimmy Lei Ba, and Ruslan Salakhutdinov. "Actor-mimic: Deep multitask and transfer reinforcement learning." arXiv preprint arXiv:1511.06342 (2015). [pdf] (RL 領域)

地址：
http://arxiv.org/pdf/1511.06342

[58] Rusu, Andrei A., et al. "Progressive neural networks." arXiv preprint arXiv:1606.04671 (2016). [pdf] (傑出研究, 很新奇的想法)

地址：
https://arxiv.org/pdf/1606.04671

2.8 One Shot 深度學習

[59] Lake, Brenden M., Ruslan Salakhutdinov, and Joshua B. Tenenbaum. "Human-level concept learning through probabilistic program induction." Science 350.6266 (2015): 1332-1338. [pdf] (不含深度學習但值得一讀)

地址：
http://clm.utexas.edu/compjclub/wp-content/uploads/2016/02/lake2015.pdf

[60] Koch, Gregory, Richard Zemel, and Ruslan Salakhutdinov. "Siamese Neural Networks for One-shot Image Recognition."(2015) [pdf]

地址：
http://www.cs.utoronto.ca/~gkoch/files/msc-thesis.pdf

[61] Santoro, Adam, et al. "One-shot Learning with Memory-Augmented Neural Networks." arXiv preprint arXiv:1605.06065 (2016). [pdf] (one shot 學習的基礎一步)

地址：
http://arxiv.org/pdf/1605.06065

[62] Vinyals, Oriol, et al. "Matching Networks for One Shot Learning." arXiv preprint arXiv:1606.04080 (2016). [pdf]

地址：
https://arxiv.org/pdf/1606.04080

[63] Hariharan, Bharath, and Ross Girshick. "Low-shot visual object recognition." arXiv preprint arXiv:1606.02819 (2016). [pdf] (通向更大規模資料的一步)

地址：
http://arxiv.org/pdf/1606.02819

3 應用

3.1 自然語言處理 (NLP)

[1] Antoine Bordes, et al. "Joint Learning of Words and Meaning Representations for Open-Text Semantic Parsing." AISTATS(2012) [pdf]

地址：
https://www.hds.utc.fr/~bordesan/dokuwiki/lib/exe/fetch.php?id=en%3Apubli&cache=cache&media=en:bordes12aistats.pdf

[2] Mikolov, et al. "Distributed representations of words and phrases and their compositionality." ANIPS(2013): 3111-3119 [pdf] (word2vec)

地址：
http://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf

[3] Sutskever, et al. "“" ANIPS(2014) [pdf]

地址：
http://papers.nips.cc/paper/5346-sequence-to-sequence-learning-with-neural-networks.pdf

[4] Ankit Kumar, et al. "“Ask Me Anything: Dynamic Memory Networks for Natural Language Processing." arXiv preprint arXiv:1506.07285(2015) [pdf]

地址：
https://arxiv.org/abs/1506.07285

[5] Yoon Kim, et al. "Character-Aware Neural Language Models." NIPS(2015) arXiv preprint arXiv:1508.06615(2015) [pdf]

地址：
https://arxiv.org/abs/1508.06615

[6] Jason Weston, et al. "Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks." arXiv preprint arXiv:1502.05698(2015) [pdf] (bAbI tasks)

地址：
https://arxiv.org/abs/1502.05698

[7] Karl Moritz Hermann, et al. "Teaching Machines to Read and Comprehend." arXiv preprint arXiv:1506.03340(2015) [pdf](CNN/每日郵報完形填空風格的問題)

地址：
https://arxiv.org/abs/1506.03340

[8] Alexis Conneau, et al. "Very Deep Convolutional Networks for Natural Language Processing." arXiv preprint arXiv:1606.01781(2016) [pdf] (文字分類的前沿技術)

地址：
https://arxiv.org/abs/1606.01781

[9] Armand Joulin, et al. "Bag of Tricks for Efficient Text Classification." arXiv preprint arXiv:1607.01759(2016) [pdf] (比前沿技術稍落後, 但快很多)

地址：
https://arxiv.org/abs/1607.01759

3.2 物體檢測

[1] Szegedy, Christian, Alexander Toshev, and Dumitru Erhan. "Deep neural networks for object detection." Advances in Neural Information Processing Systems. 2013. [pdf]

地址：
http://papers.nips.cc/paper/5207-deep-neural-networks-for-object-detection.pdf

[2] Girshick, Ross, et al. "Rich feature hierarchies for accurate object detection and semantic segmentation." Proceedings of the IEEE conference on computer vision and pattern recognition. 2014. [pdf] (RCNN)

地址：
http://www.cv-foundation.org/openaccess/content_cvpr_2014/papers/Girshick_Rich_Feature_Hierarchies_2014_CVPR_paper.pdf

[3] He, Kaiming, et al. "Spatial pyramid pooling in deep convolutional networks for visual recognition." European Conference on Computer Vision. Springer International Publishing, 2014. [pdf] (SPPNet)

地址：
http://arxiv.org/pdf/1406.4729

[4] Girshick, Ross. "Fast r-cnn." Proceedings of the IEEE International Conference on Computer Vision. 2015. [pdf]

地址：
https://pdfs.semanticscholar.org/8f67/64a59f0d17081f2a2a9d06f4ed1cdea1a0ad.pdf

[5] Ren, Shaoqing, et al. "Faster R-CNN: Towards real-time object detection with region proposal networks.

地址：
http://papers.nips.cc/paper/5638-analysis-of-variational-bayesian-latent-dirichlet-allocation-weaker-sparsity-than-map.pdf

[6] Redmon, Joseph, et al. "You only look once: Unified, real-time object detection." arXiv preprint arXiv:1506.02640 (2015). [pdf] (YOLO，傑出研究，非常具有使用價值）

地址：
http://homes.cs.washington.edu/~ali/papers/YOLO.pdf

[7] Liu, Wei, et al. "SSD: Single Shot MultiBox Detector." arXiv preprint arXiv:1512.02325 (2015). [pdf]

地址：
http://arxiv.org/pdf/1512.02325

[8] Dai, Jifeng, et al. "R-FCN: Object Detection via Region-based Fully Convolutional Networks." arXiv preprint arXiv:1605.06409 (2016). [pdf]

地址：
https://arxiv.org/abs/1605.06409

3.3 視覺追蹤

[1] Wang, Naiyan, and Dit-Yan Yeung. "Learning a deep compact image representation for visual tracking." Advances in neural information processing systems. 2013. [pdf] (第一篇使用深度學習做視覺追蹤的論文，DLT Tracker)

地址：
http://papers.nips.cc/paper/5192-learning-a-deep-compact-image-representation-for-visual-tracking.pdf

[2] Wang, Naiyan, et al. "Transferring rich feature hierarchies for robust visual tracking." arXiv preprint arXiv:1501.04587 (2015). [pdf] (SO-DLT)

地址：
http://arxiv.org/pdf/1501.04587

[3] Wang, Lijun, et al. "Visual tracking with fully convolutional networks." Proceedings of the IEEE International Conference on Computer Vision. 2015. [pdf] (FCNT)

地址：
http://www.cv-foundation.org/openaccess/content_iccv_2015/papers/Wang_Visual_Tracking_With_ICCV_2015_paper.pdf

[4] Held, David, Sebastian Thrun, and Silvio Savarese. "Learning to Track at 100 FPS with Deep Regression Networks." arXiv preprint arXiv:1604.01802 (2016). [pdf] (GOTURN，在深度學習方法裡算是非常快的，但仍比非深度學習方法慢很多)

地址：
http://arxiv.org/pdf/1604.01802

[5] Bertinetto, Luca, et al. "Fully-Convolutional Siamese Networks for Object Tracking." arXiv preprint arXiv:1606.09549 (2016). [pdf] (SiameseFC，實時物體追蹤領域的最新前沿技術)

地址：
https://arxiv.org/pdf/1606.09549

[6] Martin Danelljan, Andreas Robinson, Fahad Khan, Michael Felsberg. "Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking." ECCV (2016) [pdf] (C-COT)

地址：
http://www.cvl.isy.liu.se/research/objrec/visualtracking/conttrack/C-COT_ECCV16.pdf

[7] Nam, Hyeonseob, Mooyeol Baek, and Bohyung Han. "Modeling and Propagating CNNs in a Tree Structure for Visual Tracking." arXiv preprint arXiv:1608.07242 (2016). [pdf] (VOT2016 獲獎論文,TCNN)

地址：
https://arxiv.org/pdf/1608.07242

3.4 影像標註

[1] Farhadi,Ali,etal. "Every picture tells a story: Generating sentences from images". In Computer VisionECCV 2010. Springer Berlin Heidelberg:15-29, 2010. [pdf]

地址：
https://www.cs.cmu.edu/~afarhadi/papers/sentence.pdf

[2] Kulkarni, Girish, et al. "Baby talk: Understanding and generating image descriptions". In Proceedings of the 24th CVPR, 2011. [pdf]

地址：
http://tamaraberg.com/papers/generation_cvpr11.pdf

[3] Vinyals, Oriol, et al. "Show and tell: A neural image caption generator". In arXiv preprint arXiv:1411.4555, 2014. [pdf]

地址：
https://arxiv.org/pdf/1411.4555.pdf

[4] Donahue, Jeff, et al. "Long-term recurrent convolutional networks for visual recognition and description". In arXiv preprint arXiv:1411.4389 ,2014. [pdf]

地址：
https://arxiv.org/pdf/1411.4389.pdf

[5] Karpathy, Andrej, and Li Fei-Fei. "Deep visual-semantic alignments for generating image descriptions". In arXiv preprint arXiv:1412.2306, 2014. [pdf]

地址：
https://cs.stanford.edu/people/karpathy/cvpr2015.pdf

[6] Karpathy, Andrej, Armand Joulin, and Fei Fei F. Li. "Deep fragment embeddings for bidirectional image sentence mapping". In Advances in neural information processing systems, 2014. [pdf]

地址：
https://arxiv.org/pdf/1406.5679v1.pdf

[7] Fang, Hao, et al. "From captions to visual concepts and back". In arXiv preprint arXiv:1411.4952, 2014. [pdf]

地址：
https://arxiv.org/pdf/1411.4952v3.pdf

[8] Chen, Xinlei, and C. Lawrence Zitnick. "Learning a recurrent visual representation for image caption generation". In arXiv preprint arXiv:1411.5654, 2014. [pdf]

地址：
https://arxiv.org/pdf/1411.5654v1.pdf

[9] Mao, Junhua, et al. "Deep captioning with multimodal recurrent neural networks (m-rnn)". In arXiv preprint arXiv:1412.6632, 2014. [pdf]

地址：
https://arxiv.org/pdf/1412.6632v5.pdf

[10] Xu, Kelvin, et al. "Show, attend and tell: Neural image caption generation with visual attention". In arXiv preprint arXiv:1502.03044, 2015. [pdf]

地址：
https://arxiv.org/pdf/1502.03044v3.pdf

3.5 機器翻譯

部分里程碑研究被列入 RNN / Seq-to-Seq 版塊。

[1] Luong, Minh-Thang, et al. "Addressing the rare word problem in neural machine translation." arXiv preprint arXiv:1410.8206 (2014). [pdf]

地址：
http://arxiv.org/pdf/1410.8206

[2] Sennrich, et al."Neural Machine Translation of Rare Words with Subword Units". In arXiv preprint arXiv:1508.07909, 2015. [pdf]

地址：
https://arxiv.org/pdf/1508.07909.pdf

[3] Luong, Minh-Thang, Hieu Pham, and Christopher D. Manning. "Effective approaches to attention-based neural machine translation." arXiv preprint arXiv:1508.04025 (2015). [pdf]

地址：
http://arxiv.org/pdf/1508.04025

[4] Chung, et al. "A Character-Level Decoder without Explicit Segmentation for Neural Machine Translation". In arXiv preprint arXiv:1603.06147, 2016. [pdf]

地址：
https://arxiv.org/pdf/1603.06147.pdf

[5] Lee, et al. "Fully Character-Level Neural Machine Translation without Explicit Segmentation". In arXiv preprint arXiv:1610.03017, 2016. [pdf]

地址：
https://arxiv.org/pdf/1610.03017.pdf

[6] Wu, Schuster, Chen, Le, et al. "Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation". In arXiv preprint arXiv:1609.08144v2, 2016. [pdf] (Milestone)

地址：
https://arxiv.org/pdf/1609.08144v2.pdf

3.6 機器人

[1] Koutník, Jan, et al. "Evolving large-scale neural networks for vision-based reinforcement learning." Proceedings of the 15th annual conference on Genetic and evolutionary computation. ACM, 2013. [pdf]

地址：
http://repository.supsi.ch/4550/1/koutnik2013gecco.pdf

[2] Levine, Sergey, et al. "End-to-end training of deep visuomotor policies." Journal of Machine Learning Research 17.39 (2016): 1-40. [pdf]

地址：
http://www.jmlr.org/papers/volume17/15-522/15-522.pdf

[3] Pinto, Lerrel, and Abhinav Gupta. "Supersizing self-supervision: Learning to grasp from 50k tries and 700 robot hours." arXiv preprint arXiv:1509.06825 (2015). [pdf]

地址：
http://arxiv.org/pdf/1509.06825

[4] Levine, Sergey, et al. "Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collection." arXiv preprint arXiv:1603.02199 (2016). [pdf]

地址：
http://arxiv.org/pdf/1603.02199

[5] Zhu, Yuke, et al. "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning." arXiv preprint arXiv:1609.05143 (2016). [pdf]

地址：
https://arxiv.org/pdf/1609.05143

[6] Yahya, Ali, et al. "Collective Robot Reinforcement Learning with Distributed Asynchronous Guided Policy Search." arXiv preprint arXiv:1610.00673 (2016). [pdf]

地址：
https://arxiv.org/pdf/1610.00673

[7] Gu, Shixiang, et al. "Deep Reinforcement Learning for Robotic Manipulation." arXiv preprint arXiv:1610.00633 (2016). [pdf]

地址：
https://arxiv.org/pdf/1610.00633

[8] A Rusu, M Vecerik, Thomas Rothrl, N Heess, R Pascanu, R Hadsell."Sim-to-Real Robot Learning from Pixels with Progressive Nets." arXiv preprint arXiv:1610.04286 (2016). [pdf]

地址：
https://arxiv.org/pdf/1610.04286.pdf

[9] Mirowski, Piotr, et al. "Learning to navigate in complex environments." arXiv preprint arXiv:1611.03673 (2016). [pdf]

地址：
https://arxiv.org/pdf/1611.03673

3.7 藝術

[1] Mordvintsev, Alexander; Olah, Christopher; Tyka, Mike (2015). "Inceptionism: Going Deeper into Neural Networks". Google Research. [html] (Deep Dream)

地址：
https://research.googleblog.com/2015/06/inceptionism-going-deeper-into-neural.html

[2] Gatys, Leon A., Alexander S. Ecker, and Matthias Bethge. "A neural algorithm of artistic style." arXiv preprint arXiv:1508.06576 (2015). [pdf] (傑出研究，迄今最成功的方法)

地址：
http://arxiv.org/pdf/1508.06576

[3] Zhu, Jun-Yan, et al. "Generative Visual Manipulation on the Natural Image Manifold." European Conference on Computer Vision. Springer International Publishing, 2016. [pdf] (iGAN)

地址：
https://arxiv.org/pdf/1609.03552

[4] Champandard, Alex J. "Semantic Style Transfer and Turning Two-Bit Doodles into Fine Artworks." arXiv preprint arXiv:1603.01768 (2016). [pdf] (Neural Doodle)

地址：
http://arxiv.org/pdf/1603.01768

[5] Zhang, Richard, Phillip Isola, and Alexei A. Efros. "Colorful Image Colorization." arXiv preprint arXiv:1603.08511 (2016). [pdf]

地址：
http://arxiv.org/pdf/1603.08511

[6] Johnson, Justin, Alexandre Alahi, and Li Fei-Fei. "Perceptual losses for real-time style transfer and super-resolution." arXiv preprint arXiv:1603.08155 (2016). [pdf]

地址：
https://arxiv.org/pdf/1603.08155.pdf

[7] Vincent Dumoulin, Jonathon Shlens and Manjunath Kudlur. "A learned representation for artistic style." arXiv preprint arXiv:1610.07629 (2016). [pdf]

[8] Gatys, Leon and Ecker, et al."Controlling Perceptual Factors in Neural Style Transfer." arXiv preprint arXiv:1611.07865 (2016). [pdf] (control style transfer over spatial location,colour information and across spatial scale)

[9] Ulyanov, Dmitry and Lebedev, Vadim, et al. "Texture Networks: Feed-forward Synthesis of Textures and Stylized Images." arXiv preprint arXiv:1603.03417(2016). [pdf] (紋理生成和風格變化)

3.8 目標分割 Object Segmentation

[1] J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation.” in CVPR, 2015. [pdf]

地址：
https://arxiv.org/pdf/1411.4038v2.pdf

[2] L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille. "Semantic image segmentation with deep convolutional nets and fully connected crfs." In ICLR, 2015. [pdf]

地址：
https://arxiv.org/pdf/1606.00915v1.pdf

[3] Pinheiro, P.O., Collobert, R., Dollar, P. "Learning to segment object candidates." In: NIPS. 2015. [pdf]

地址：
https://arxiv.org/pdf/1506.06204v2.pdf

[4] Dai, J., He, K., Sun, J."Instance-aware semantic segmentation via multi-task network cascades." in CVPR. 2016 [pdf]

地址：
https://arxiv.org/pdf/1512.04412v1.pdf

[5] Dai, J., He, K., Sun, J. "Instance-sensitive Fully Convolutional Networks." arXiv preprint arXiv:1603.08678 (2016). [pdf]

地址：
https://arxiv.org/pdf/1603.08678v1.pdf