-
CNN
- http://mclab.eic.hust.edu.cn/valse2016/dl/刘昕.pdf
- http://chuansong.me/n/316919551459
- http://chuansong.me/n/320583251483
- http://chuansong.me/n/322938451654
- http://arxiv.org/abs/1404.5997
- http://arxiv.org/abs/1603.07285
- https://github.com/vdumoulin/conv_arithmetic
- http://arxiv.org/abs/1506.03059
- http://cogprints.org/5869/1/cnn_tutorial.pdf
- http://arxiv.org/abs/1606.02228
- https://github.com/ducha-aiki/caffenet-benchmark
- https://arxiv.org/abs/1601.04920
- https://github.com/Tetrachrome/subpixel
- https://arxiv.org/abs/1609.05158
- https://arxiv.org/abs/1311.2901
- https://arxiv.org/abs/1609.07009
- https://arxiv.org/abs/1312.5402
- http://distill.pub/2016/deconv-checkerboard/
- https://arxiv.org/abs/1611.00847
- https://arxiv.org/abs/1412.6806
- http://arxiv.org/abs/1511.07571
- http://arxiv.org/abs/1512.07108
- http://arxiv.org/abs/1604.01792
- http://arxiv.org/abs/1601.06615
- http://lamda.nju.edu.cn/weixs/project/CNNTricks/CNNTricks.html
- http://lamda.nju.edu.cn/weixs/slide/CNNTricks_slide.pdf
- https://medium.com/towards-data-science/linear-algebra-cheat-sheet-for-deep-learning-cd67aba4526c
- https://arxiv.org/abs/1703.06817
- https://arxiv.org/abs/1506.02025
- https://arxiv.org/abs/1703.06211
- https://arxiv.org/abs/1701.04128
- https://medium.com/@nikasa1889/a-guide-to-receptive-field-arithmetic-for-convolutional-neural-networks-e0f514068807
- http://www.dongzhuoyao.com/convolution-understanding/
- https://arxiv.org/abs/1509.05009
- https://arxiv.org/abs/1511.07122
- https://arxiv.org/abs/1511.07356
- https://arxiv.org/abs/1512.09194
- https://arxiv.org/abs/1703.02065
- https://arxiv.org/abs/1703.06846
- https://www.zhihu.com/question/43609045
- https://arxiv.org/abs/1605.02971
- https://arxiv.org/abs/1609.04112
- keras-team/keras#1777
- https://github.com/Lasagne/Recipes/blob/master/examples/Saliency%20Maps%20and%20Guided%20Backpropagation.ipynb
- https://github.com/kundajelab/deeplift/blob/tensorflow/examples/public/tal_gata/TAL-GATA%20simulation.ipynb
- http://stackoverflow.com/questions/38340791/guided-back-propagation-in-tensorflow
- https://arxiv.org/abs/1412.6806
- http://www.cs.toronto.edu/~guerzhoy/321/lec/W07/HowConvNetsSee.pdf
- tensorflow/tensorflow#6422
- http://stackoverflow.com/questions/39793505/in-tensorflow-is-it-possible-to-use-different-learning-rate-for-different-part/39793644#39793644
- http://stackoverflow.com/questions/35298326/freeze-some-variables-scopes-in-tensorflow-stop-gradient-vs-passing-variables
- https://arxiv.org/abs/1608.00530
- https://github.com/artvandelay/Deep_Inside_Convolutional_Networks
- https://arxiv.org/abs/1611.05418
- https://arxiv.org/abs/1704.07911
- http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7578651
- http://web.mit.edu/zoya/www/visual_attention_reading.pdf
- https://github.com/kjw0612/awesome-deep-vision#visual-attention-and-saliency
- http://www.cse.ust.hk/~hzengac/PQE/survey_for_pqe.pdf
- https://arxiv.org/abs/1604.07043
- https://arxiv.org/abs/1506.06579
- https://arxiv.org/abs/1602.03616
- https://chenzomi12.github.io/2016/12/13/CNN-Architectures/
- https://arxiv.org/abs/1605.07678
- https://arxiv.org/abs/1705.01450
- https://github.com/ankurtaly/Attributions
- https://zhuanlan.zhihu.com/c_162633442
- https://towardsdatascience.com/convolutional-neural-networks-from-the-ground-up-c67bb41454e1
- https://wiseodd.github.io/techblog/2016/07/16/convnet-conv-layer/
- https://distill.pub/2016/deconv-checkerboard/
- https://towardsdatascience.com/how-to-visualize-convolutional-features-in-40-lines-of-code-70b7d87b0030
-
人脸
- http://mftp.mmcheng.net/Reports
- http://chuansong.me/n/290740551733
- http://chuansong.me/n/290740251267
- http://chuansong.me/n/290740151865
- https://arxiv.org/abs/1606.03073
- http://chuansong.me/n/466689951179
- http://chuansong.me/n/436893551836
- https://zo7.github.io/blog/2016/09/25/generating-faces.html
- https://github.com/zo7/deconvfaces
- https://arxiv.org/abs/1701.07174
- https://github.com/ShiqiYu/libfacedetection
- https://github.com/qiexing/face-landmark-localization
- http://cmusatyalab.github.io/openface/
-
加速
-
http://on-demand.gputechconf.com/gtc/2016/presentation/s6673-greg-diamos-persisten-rnns.pdf
-
http://archnews.blogspot.si/2015/11/accelerating-watsons-performance-with.html
-
https://blogs.nvidia.com/blog/2016/01/12/accelerating-ai-artificial-intelligence-gpus/
-
http://cnblog.lingcc.com/2016/05/20/Google-TPU-Machine-Learning-Accelerator/
-
http://www.fahim-kawsar.net/papers/Lane.HotMobile2016-Camera.pdf
-
https://research.facebook.com/blog/fair-open-sources-deep-learning-modules-for-torch/
-
https://devblogs.nvidia.com/parallelforall/optimizing-recurrent-neural-networks-cudnn-5/
-
http://mi.eng.cam.ac.uk/projects/cued-rnnlm/papers/RNNLMTrain_Interspeech2014.pdf
-
http://uploads.movidius.com/1441734401-Myriad-2-product-brief.pdf
-
http://pages.saclay.inria.fr/olivier.temam/files/eval/DLCPTW2014.pdf
-
http://pages.saclay.inria.fr/olivier.temam/files/eval/CDSWWCT14.pdf
-
http://courses.cs.washington.edu/courses/cse590g/16wi/du_micro15.pdf
-
http://static.googleusercontent.com/media/research.google.com/zh-CN//pubs/archive/37631.pdf
-
http://research.microsoft.com/pubs/240715/CNN%20Whitepaper.pdf
-
http://www.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-53.pdf
-
http://on-demand.gputechconf.com/gtc/2015/webinar/deep-learning-course/intro-to-deep-learning.pdf
-
http://www.nextplatform.com/2016/06/29/universal-fpga-gpu-platform-deep-learning/
-
http://www.nextplatform.com/2016/06/21/knights-landing-solid-ground-intels-stake-deep-learning/
-
http://www.nextplatform.com/2016/06/20/intel-knights-landing-yields-big-bang-buck-jump/
-
http://ip.cadence.com/uploads/presentations/1100AM_TensorFlow_on_Embedded_Devices_PeteWarden.pdf
-
https://www.codeplay.com/portal/tensorflow™-for-opencl™-using-sycl™
-
http://www.ece.ucdavis.edu/~soheil/publications/conference/DCNN_16.pdf
-
http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7424676
-
http://www.tomshardware.com/news/embedded-client-chips-deep-learning,31775.html
-
https://devblogs.nvidia.com/parallelforall/inference-next-step-gpu-accelerated-deep-learning/
-
https://www.nvidia.com/content/tegra/embedded-systems/pdf/jetson_tx1_whitepaper.pdf
-
https://www.ibr.cs.tu-bs.de/Cosdeo2016/talks/invitedTalk.pdf
-
http://www.nextplatform.com/2016/08/08/deep-learning-chip-upstart-set-take-gpus-task/
-
http://www.nextplatform.com/2016/08/10/nervana-ceo-intel-acquisition-future-technology-outlook/
-
http://www.nextbigfuture.com/2016/08/startup-nervana-making-deep-learning.html
-
http://stackoverflow.com/questions/4097034/arm-cortex-a8-whats-the-difference-between-vfp-and-neon
-
https://www.quora.com/What-is-the-difference-between-mali-and-adreno-GPU-in-Android
-
http://www.iwocl.org/wp-content/uploads/iwocl-2015-talk-Eric-Demers-Qualcomm.pdf
-
https://imgtec.com/blog/deep-learning-mobile-embedded-vision-summit/
-
http://www.nextplatform.com/2016/08/22/intel-tweaking-xeon-phi-deep-learning/
-
http://www.nextplatform.com/2016/09/01/cpu-gpu-put-deep-learning-framework-test/
-
http://www.nextplatform.com/2016/09/08/refreshed-ibm-power-linux-systems-add-nvlink/
-
http://www.rle.mit.edu/eems/wp-content/uploads/2016/04/eyeriss_isca_2016.pdf
-
http://people.csail.mit.edu/emer/slides/2016.02.isscc.eyeriss.slides.pdf
-
http://www.knowledgegrid.net/skg2016/keynotePPT/SKG-2016-ZWXu.pdf
-
https://media.nips.cc/Conferences/2015/tutorialslides/Dally-NIPS-Tutorial-2015.pdf
-
http://www.isfpga.org/fpga2016/index_files/FPGA2016FinalProgram2_withSlides.htm
-
https://code.facebook.com/posts/196146247499076/delivering-real-time-ai-in-the-palm-of-your-hand/
-
https://www.nextplatform.com/2017/02/02/memory-core-new-deep-learning-research-chip/
-
https://www.nextplatform.com/2017/04/05/first-depth-look-googles-tpu-architecture/
-
https://drive.google.com/file/d/0Bx4hafXDDq2EMzRNcy1vSUxtcEk/view
-
LowPrecision
- https://petewarden.com/2015/05/23/why-are-eight-bits-enough-for-deep-neural-networks/
- http://jmlr.org/proceedings/papers/v37/gupta15.pdf
- http://www.labs.hpe.com/techreports/2015/HPL-2015-100.pdf
- https://arxiv.org/abs/1502.02551
- http://bbs.ednchina.com/BLOG_ARTICLE_3013461.HTM
- http://blog.csdn.net/jackytintin/article/details/7798157
- http://arxiv.org/abs/1412.7024
- http://arxiv.org/abs/1511.04561
- http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6986082
- http://arxiv.org/abs/1510.03009
- http://arxiv.org/abs/1511.00363
- http://arxiv.org/abs/1602.02830
- https://github.com/codekansas/tinier-nn
- https://github.com/MatthieuCourbariaux/BinaryConnect
- https://github.com/hantek/BinaryConnect
- http://arxiv.org/abs/1601.06071
- http://arxiv.org/abs/1603.05279
- https://arxiv.org/abs/1605.04711
- http://arxiv.org/abs/1609.00222
- https://arxiv.org/abs/1606.05487
- http://arxiv.org/abs/1606.06160
- https://github.com/ppwwyyxx/tensorpack/tree/master/examples/DoReFa-Net
- https://petewarden.com/2016/05/03/how-to-quantize-neural-networks-with-tensorflow/
- http://arxiv.org/abs/1412.6115
- http://research.microsoft.com/apps/pubs/?id=230137
- http://research.microsoft.com/pubs/230137/IS140694.PDF
- http://papers.nips.cc/paper/5787-tensorizing-neural-networks.pdf
- http://www.nikkostrom.com/publications/interspeech2015/strom_interspeech2015.pdf
- https://github.com/NervanaSystems/nervana-lib-gpu-performance-preview
- https://arxiv.org/abs/1601.07789
- http://cas.ee.ic.ac.uk/people/gac1/pubs/Umar14.pdf
- http://arxiv.org/abs/1512.06473
- https://github.com/jiaxiang-wu/quantized-cnn
- tiny-dnn/tiny-dnn#202
- https://arxiv.org/abs/1604.03168
- https://arxiv.org/abs/1605.06402
- http://ristretto.lepsucd.com/
- https://github.com/pmgysel/caffe
- tensorflow/tensorflow#1300
- http://arxiv.org/abs/1504.07488
- https://arxiv.org/abs/1607.02241
- https://arxiv.org/abs/1605.06402
- http://arxiv.org/abs/1603.06777
- https://arxiv.org/abs/1511.06744
- https://nicsefc.ee.tsinghua.edu.cn/media/publications/2016/ISCAS16_194.pdf
- https://nicsefc.ee.tsinghua.edu.cn/media/publications/2016/ISCAS16_194_slide.pdf
- https://arxiv.org/abs/1608.06902
- http://www.voidcn.com/blog/CHIERYU/article/p-4928473.html
- http://www.voidcn.com/blog/CHIERYU/article/p-4928475.html
- http://www.voidcn.com/blog/CHIERYU/article/p-4928477.html
- http://www.voidcn.com/blog/CHIERYU/article/p-4943264.html
- http://www.cs.ubc.ca/~julm/papers/eccv16.pdf
- http://arxiv.org/abs/1606.01981
- https://github.com/ahmetaa/fast-dnn
- http://arxiv.org/abs/1512.06473
- http://arxiv.org/abs/1511.06393
- http://www.prism.gatech.edu/~tna6/islped2016.pdf
- http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7178146
- https://nickhigham.wordpress.com/2015/10/20/the-rise-of-mixed-precision-arithmetic/
- https://arxiv.org/abs/0808.2794
- http://digitalassets.lib.berkeley.edu/techreports/ucb/text/EECS-2015-6.pdf
- https://www.nextplatform.com/2017/01/26/fpgas-focal-point-efficient-neural-network-inference/
- https://www.nervanasys.com/accelerating-neural-networks-binary-arithmetic/
- https://arxiv.org/abs/1701.03400
- http://arainhyy.github.io/
- https://arxiv.org/abs/1612.07119
- https://arxiv.org/abs/1609.07061
-
压缩
- http://arxiv.org/abs/1511.05641
- http://arxiv.org/abs/1512.01891
- http://arxiv.org/abs/1511.06530
- http://arxiv.org/abs/1512.08571
- http://handong1587.github.io/deep_learning/2015/10/09/acceleration_model_compression.html
- http://blog.csdn.net/shuzfan/article/category/6271575
- http://arxiv.org/abs/1608.03665
- https://github.com/wenwei202/caffe/tree/scnn
- http://arxiv.org/abs/1507.06149
- http://arxiv.org/abs/1606.09274
- http://nlp.stanford.edu/courses/cs224n/2015/reports/26.pdf
- http://www.ijcai.org/Proceedings/16/Papers/251.pdf
- http://cs231n.stanford.edu/reports2016/117_Report.pdf
- http://www.cv-foundation.org/openaccess/content_iccv_2015/papers/Yang_Deep_Fried_Convnets_ICCV_2015_paper.pdf
- https://arxiv.org/abs/1412.7149
- http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Liu_Sparse_Convolutional_Neural_2015_CVPR_paper.pdf
- http://arxiv.org/abs/1604.06832
- http://arxiv.org/abs/1506.04449
- http://papers.nips.cc/paper/5869-structured-transforms-for-small-footprint-deep-learning.pdf
- http://arxiv.org/abs/1605.06560
- http://arxiv.org/abs/1504.04788
- http://arxiv.org/abs/1408.3060
- https://arxiv.org/abs/1306.0543
- https://arxiv.org/abs/1411.4229
- http://cvlab.postech.ac.kr/~bhhan/class/cse703r_2016s/Mooyeol_CNN_Compression.pptx
- https://arxiv.org/abs/1606.05426
- https://arxiv.org/abs/1501.00777
- http://static.googleusercontent.com/media/research.google.com/zh-CN//pubs/archive/41176.pdf
- http://arxiv.org/abs/1608.01409
- http://nlp.stanford.edu/courses/cs224n/2015/reports/2.pdf
- http://www.kdnuggets.com/2015/05/dark-knowledge-neural-network.html
- http://www.ttic.edu/dl/dark14.pdf
- https://github.com/yuanyuanli85/CaffeModelCompression
- http://www.kdd.org/kdd2016/papers/files/rpp0534-chenA.pdf
- https://arxiv.org/abs/1506.03767
- http://arxiv.org/abs/1509.08745
- http://openscholarship.wustl.edu/eng_etds/155/
- https://www.microsoft.com/en-us/research/publication/exploiting-sparseness-in-deep-neural-networks-for-large-vocabulary-speech-recognition/
- https://arxiv.org/abs/1606.02147
- http://people.idsia.ch/~juergen/gecco2010koutnik.pdf
- https://arxiv.org/abs/1606.06216
- https://arxiv.org/abs/1701.04923
- https://arxiv.org/abs/1702.04008
-
HanSong
-
http://www.nextplatform.com/2015/12/08/emergent-chip-vastly-accelerates-deep-neural-networks/
-
http://on-demand.gputechconf.com/gtc/2016/presentation/s6561-song-han-deep-compression.pdf
-
http://web.stanford.edu/class/ee380/Abstracts/160106-slides.pdf
-
DeepCompression.net
-
Username: deepcompression
-
Password: songhan
-
http://openpowerfoundation.org/wp-content/uploads/2016/02/12_Ralph-Wittig.Xilinx.pdf
-
https://nicsefc.ee.tsinghua.edu.cn/media/publications/2016/FPGA2016_None_6tAJnDW.pdf
-
https://nicsefc.ee.tsinghua.edu.cn/media/publications/2016/FPGA2016_None_slide.pdf
-
http://www.nextplatform.com/2016/08/23/fpga-based-deep-learning-accelerators-take-asics/
-
-
DivNet