机器学习/计算机视觉/ NLP的论文及笔记

【导读】这个仓库包含了一些机器学习论文与笔记,主题包括自监督学习、半监督学习、无监督学习、语义分割、弱监督、半监督语义分割、信息检索、图神经网络等。

原文链接:https://github.com/yassouali/ML_paper_notes

Self-Supervised Learning

  • Selfie: Self-supervised Pretraining for Image Embedding (2019):
    • Paper https://arxiv.org/abs/1906.02940
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/76_selfie_pretraining_for_img_embeddings.pdf
  • Self-Supervised Representation Learning by Rotation Feature Decoupling (2019):
    • Paper https://github.com/philiptheother/FeatureDecoupling
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/73_SSL_by_rotation_decoupling.pdf
  • Revisiting Self-Supervised Visual Representation Learning (2019):
    • Paper https://arxiv.org/abs/1901.09005
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/72_revisiting_SSL.pdf
  • AET vs. AED: Unsupervised Representation Learning by Auto-Encoding Transformations rather than Data (2019):
    • Paper https://arxiv.org/abs/1901.04596
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/74_AFT_vs_AED.pdf
  • Boosting Self-Supervised Learning via Knowledge Transfer (2018):
    • Paper https://arxiv.org/abs/1805.00385
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/67_boosting_self_super_via_trsf_learning.pdf
  • Self-Supervised Feature Learning by Learning to Spot Artifacts (2018):
    • Paper https://arxiv.org/abs/1806.05024
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/69_SSL_by_learn_to_spot_artifacts.pdf
  • Unsupervised Representation Learning by Predicting Image Rotations (2018):
    • Paper https://arxiv.org/abs/1803.07728
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/68_unsup_img_rep_learn_by_rot_predic.pdf
  • Cross Pixel Optical-Flow Similarity for Self-Supervised Learning (2018):
    • Paper https://arxiv.org/abs/1807.05636
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/75_cross_pixel_optical_flow.pdf
  • Multi-task Self-Supervised Visual Learning (2017):
    • Paper https://arxiv.org/abs/1708.07860
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/64_multi_task_self_supervised.pdf
  • Split-Brain Autoencoders: Unsupervised Learning by Cross-Channel Prediction (2017):
    • Paper https://arxiv.org/abs/1611.09842
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/65_split_brain_autoencoders.pdf
  • Colorization as a Proxy Task for Visual Understanding (2017):
    • Paper https://arxiv.org/abs/1703.04044
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/66_colorization_as_a_proxy_for_viz_under.pdf
  • Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzles (2017):
    • Paper https://arxiv.org/abs/1603.09246
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/63_solving_jigsaw_puzzles.pdf
  • Unsupervised Visual Representation Learning by Context Prediction (2016):
    • Paper https://arxiv.org/abs/1505.05192
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/62_unsupervised_learning_with_context_prediction.pdf
  • Colorful image colorization (2016):
    • Paper https://richzhang.github.io/colorization/
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/59_colorful_colorization.pdf
  • Learning visual groups from co-occurrences in space and time (2015):
    • Paper https://arxiv.org/abs/1511.06811
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/61_visual_groups_from_co_occurrences.pdf
  • Discriminative unsupervised feature learning with exemplar convolutional neural networks (2015):
    • Paper https://arxiv.org/abs/1406.6909
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/60_exemplar_CNNs.pdf

Semi-Supervised Learning

  • Dual Student: Breaking the Limits of the Teacher in Semi-supervised Learning (2019):
    • Paper https://arxiv.org/abs/1909.01804
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/79_dual_student.pdf
  • S4L: Self-Supervised Semi-Supervised Learning (2019):
    • Paper https://arxiv.org/abs/1905.03670
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/83_S4L.pdf
  • Semi-Supervised Learning by Augmented Distribution Alignment (2019
    • Paper https://arxiv.org/abs/1905.08171
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/80_SSL_aug_dist_align.pdf
  • MixMatch: A Holistic Approach toSemi-Supervised Learning (2019):
    • Paper https://arxiv.org/abs/1905.02249
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/45_mixmatch.pdf
  • Unsupervised Data Augmentation (2019):
    • Paper https://arxiv.org/abs/1904.12848
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/39_unsupervised_data_aug.pdf
  • Interpolation Consistency Training forSemi-Supervised Learning (2019):
    • Paper https://arxiv.org/abs/1903.03825
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/44_interpolation_consistency_tranining.pdf
  • Deep Co-Training for Semi-Supervised Image Recognition (2018):
    • Paper https://arxiv.org/abs/1803.05984
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/46_deep_co_training_img_rec.pdf
  • Unifying semi-supervised and robust learning by mixup (2019):
    • Paper https://openreview.net/forum?id=r1gp1jRN_4
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/42_mixmixup.pdf
  • Realistic Evaluation of Deep Semi-Supervised Learning Algorithms (2018):
    • Paper https://arxiv.org/abs/1804.09170
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/37_realistic_eval_of_deep_ss.pdf
  • Semi-Supervised Sequence Modeling with Cross-View Training (2018):
    • Paper https://arxiv.org/abs/1809.08370
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/38_cross_view_semi_supervised.pdf
  • Virtual Adversarial Training:A Regularization Method for Supervised andSemi-Supervised Learning (2017):
    • Paper https://arxiv.org/abs/1704.03976
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/40_virtual_adversarial_training.pdf
  • Mean teachers are better role models (2017):
    • Paper https://arxiv.org/abs/1703.01780
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/56_mean_teachers.pdf
  • Temporal Ensembling for Semi-Supervised Learning (2017):
    • Paper https://arxiv.org/abs/1610.02242
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/55_temporal-ensambling.pdf
  • Semi-Supervised Learning with Ladder Networks (2015):
    • Paper https://arxiv.org/abs/1507.02672
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/33_ladder_nets.pdf

Unsupervised Learning

  • Invariant Information Clustering for Unsupervised Image Classification and Segmentation (2019):
    • Paper https://arxiv.org/abs/1807.06653
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/78_IIC.pdf
  • Deep Clustering for Unsupervised Learning of Visual Feature (2018):
    • Paper https://arxiv.org/abs/1807.05520
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/70_deep_clustering_for_un_visual_features.pdf

Semantic Segmentation

  • DeepLabv3+: Encoder-Decoder with Atrous Separable Convolution (2018):
    • Paper https://arxiv.org/abs/1802.02611
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/26_deeplabv3+.pdf
  • Large Kernel Matter, Improve Semantic Segmentation by Global Convolutional Network (2017):
    • Paper https://arxiv.org/abs/1703.02719
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/28_large_kernel_maters.pdf
  • Understanding Convolution for Semantic Segmentation (2018):
    • Paper https://arxiv.org/abs/1702.08502
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/29_understanding_conv_for_sem_seg.pdf
  • Rethinking Atrous Convolution for Semantic Image Segmentation (2017):
    • Paper https://arxiv.org/abs/1706.05587
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/25_deeplab_v3.pdf
  • RefineNet: Multi-path refinement networks for high-resolution semantic segmentation (2017):
    • Paper https://arxiv.org/abs/1611.06612
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/31_refinenet.pdf
  • Pyramid Scene Parsing Network (2017):
    • Paper http://jiaya.me/papers/PSPNet_cvpr17.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/22_pspnet.pdf
  • SegNet: A Deep ConvolutionalEncoder-Decoder Architecture for ImageSegmentation (2016):
    • Paper https://arxiv.org/pdf/1511.00561
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/21_segnet.pdf
  • ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation (2016):
    • Paper https://arxiv.org/abs/1606.02147
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/27_enet.pdf
  • Attention to Scale: Scale-aware Semantic Image Segmentation (2016):
    • Paper https://arxiv.org/abs/1511.03339
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/30_atttention_to_scale.pdf
  • Deeplab: semantic image segmentation with DCNN, atrous convs and CRFs (2016):
    • Paper https://arxiv.org/abs/1606.00915
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/23_deeplab_v2.pdf
  • U-Net: Convolutional Networks for Biomedical Image Segmentation (2015):
    • Paper https://arxiv.org/abs/1505.04597
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/20_Unet.pdf
  • Fully Convolutional Networks for Semantic Segmentation (2015):
    • Paper https://people.eecs.berkeley.edu/~jonlong/long_shelhamer_fcn.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/19_FCN.pdf
  • Hypercolumns for object segmentation and fine-grained localization (2015):
    • Paper http://home.bharathh.info/pubs/pdfs/BharathCVPR2015.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/24_hypercolumns.pdf

Weakly- and Semi-supervised Semantic segmentation

  • Box-driven Class-wise Region Masking and Filling Rate Guided Loss for Weakly Supervised Semantic Segmentation (2019):
    • Paper http://arxiv.org/abs/1904.11693
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/54_boxe_driven_weakly_segmentation.pdf
  • FickleNet: Weakly and Semi-supervised Semantic Image Segmentation using Stochastic Inference (2019):
    • Paper https://arxiv.org/abs/1902.10421
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/49_ficklenet.pdf
  • Weakly-Supervised Semantic Segmentation Network with Deep Seeded Region Growing (2018):
    • Paper http://openaccess.thecvf.com/content_cvpr_2018/papers/Huang_Weakly-Supervised_Semantic_Segmentation_CVPR_2018_paper.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/53_deep_seeded_region_growing.pdf
  • Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation (2018):
    • Paper https://arxiv.org/abs/1803.10464
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/81_affinity_for_ws_segmentation.pdf
  • Object Region Mining with Adversarial Erasing: A Simple Classification to Semantic Segmentation Approach (2018):
    • Paper https://arxiv.org/abs/1703.08448
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/51_object_region_manning_for_sem_seg.pdf
  • Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation (2018):
    • Paper https://arxiv.org/abs/1805.04574
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/52_dilates_convolution_semi_super_segmentation.pdf
  • Tell Me Where to Look: Guided Attention Inference Network (2018):
    • Paper https://arxiv.org/abs/1802.10171
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/50_tell_me_where_to_look.pdf
  • Semi Supervised Semantic Segmentation Using Generative Adversarial Network (2017):
    • Paper https://arxiv.org/abs/1703.09695
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/82_ss_segmentation_gans.pdf
  • Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation (2015):
    • Paper https://arxiv.org/abs/1506.04924
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/47_decoupled_nn_for_segmentation.pdf
  • Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentation (2015):
    • Paper https://arxiv.org/abs/1502.02734
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/48_weakly_and_ss_for_segmentation.pdf

Information Retrieval

  • VSE++: Improving Visual-Semantic Embeddings with Hard Negatives (2018):
    • Paper https://arxiv.org/abs/1707.05612
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/77_vse++.pdf

Visual Explanation & Attention

  • Attention Branch Network: Learning of Attention Mechanism for Visual Explanation (2019):
    • Paper https://arxiv.org/abs/1812.10025
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/57_attention_branch_netwrok.pdf
  • Attention-based Dropout Layer for Weakly Supervised Object Localization (2019):
    • Paper http://openaccess.thecvf.com/content_CVPR_2019/papers/Choe_Attention-Based_Dropout_Layer_for_Weakly_Supervised_Object_Localization_CVPR_2019_paper.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/58_attention_based_dropout.pdf
  • Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer (2016):
    • Paper https://arxiv.org/abs/1612.03928
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/71_attention_transfer.pdf

Graph neural network & Graph embeddings

  • Pixels to Graphs by Associative Embedding (2017):
    • Paper https://arxiv.org/abs/1706.07365
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/36_pixels_to_graphs.pdf
  • Associative Embedding: End-to-End Learning forJoint Detection and Grouping (2017):
    • Paper https://arxiv.org/abs/1611.05424
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/35_associative_emb.pdf
  • Interaction Networks for Learning about Objects , Relations and Physics (2016):
    • Paper https://arxiv.org/abs/1612.00222
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/18_interaction_nets.pdf
  • DeepWalk: Online Learning of Social Representation (2014):
    • Paper http://www.perozzi.net/publications/14_kdd_deepwalk.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/deep_walk.pdf
  • The graph neural network model (2009):
    • Paper https://persagen.com/files/misc/scarselli2009graph.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/graph_neural_nets.pdf

Regularization

  • Manifold Mixup: Better Representations by Interpolating Hidden States (2018):
    • Paper https://arxiv.org/abs/1806.05236
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/43_manifold_mixup.pdf

Deep learning Methods & Models

  • AutoAugment (2018):
    • Paper https://arxiv.org/abs/1805.09501
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/41_autoaugment.pdf
  • Stacked Hourgloass (2017):
    • Paper http://ismir2018.ircam.fr/doc/pdfs/138_Paper.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/34_stacked_hourglass.pdf

Document analysis and segmentation

  • dhSegment: A generic deep-learning approach for document segmentation (2018):
    • Paper https://arxiv.org/abs/1804.10371
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/dhSegement.pdf
  • Learning to extract semantic structure from documents using multimodal fully convolutional neural networks (2017):
    • Paper https://arxiv.org/abs/1706.02337
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/learning_to_extract.pdf
  • Page Segmentation for Historical Handwritten Document Images Using Conditional Random Fields (2016):
    • Paper https://www.researchgate.net/publication/312486501_Page_Segmentation_for_Historical_Handwritten_Document_Images_Using_Conditional_Random_Fields
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/seg_with_CRFs.pdf
  • ICDAR 2015 competition on text line detection in historical documents (2015):
    • Paper http://ieeexplore.ieee.org/abstract/document/7333945/
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/ICDAR2015.pdf
  • Handwritten text line segmentation using Fully Convolutional Network (2017):
    • Paper https://ieeexplore.ieee.org/document/8270267/
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/handwritten_text_seg_FCN.pdf
  • Deep Neural Networks for Large Vocabulary Handwritten Text Recognition (2015):
    • Paper https://tel.archives-ouvertes.fr/tel-01249405/document
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/andwriten_text_recognition.pdf
  • Page Segmentation of Historical Document Images with Convolutional Autoencoders (2015):
    • Paper https://ieeexplore.ieee.org/abstract/document/7333914/
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/segmentation_with_CAE.pdf
  • A typed and handwritten text block segmentation system for heterogeneous and complex documents (2012):
    • Paper https://www.researchgate.net/publication/275518176_A_Typed_and_Handwritten_Text_Block_Segmentation_System_for_Heterogeneous_and_Complex_Documents
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/a_typed_block_seg.pdf
  • Document layout analysis, Classical approaches (1992:2001):
    • Paper https://pdfs.semanticscholar.org/5392/90b571b918da959fabaae7f605bb07850518.pdf
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/old_classical_approaches.pdf
  • Page Segmentation for Historical Document Images Based on Superpixel Classification with Unsupervised Feature Learning (2016):
    • Paper https://ieeexplore.ieee.org/document/7490134
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/seg_with_superpixels.pdf
  • Paragraph text segmentation into lines with Recurrent Neural Networks (2015):
    • Paper http://ieeexplore.ieee.org/abstract/document/7333803/
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/textlines_srg_with_RNNs.pdf
  • A comprehensive survey of mostly textual document segmentation algorithms since 2008 (2017 ):
    • Paper https://hal.archives-ouvertes.fr/hal-01388088/document
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/survey_doc_segmentation.pdf
  • Convolutional Neural Networks for Page Segmentation of Historical Document Images (2017):
    • Paper  https://arxiv.org/abs/1704.01474
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/CNNs_chen.pdf
  • ICDAR2009 Page Segmentation Competition (2009):
    • Paper https://ieeexplore.ieee.org/document/5277763
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/ICDAR2009.pdf
  • Amethod for combining complementary techniques for document image segmentation (2009):
    • Paper https://www.researchgate.net/publication/220600948_A_method_for_combining_complementary_techniques_for_document_image_segmentation
    • Notes https://github.com/yassouali/ML_paper_notes/blob/master/notes/a_method_for_combining_complementary_techniques.pdf
-END-
专 · 知


专知,专业可信的人工智能知识分发,让认知协作更快更好!欢迎注册登录专知www.zhuanzhi.ai,获取5000+AI主题干货知识资料!
欢迎微信扫一扫加入专知人工智能知识星球群,获取最新AI专业干货知识教程视频资料和与专家交流咨询
请加专知小助手微信(扫一扫如下二维码添加),获取专知VIP会员码,加入专知人工智能主题群,咨询技术商务合作~
点击“阅读原文”,了解成为 专知会员 ,查看5000+AI主题知识资料
展开全文
Top
微信扫码咨询专知VIP会员