最高一萬(wàn)星!Github標(biāo)星最多的40篇ICLR2020計(jì)算機(jī)視覺(jué)開(kāi)源論文合集,附打包下載
本文為極市平臺(tái)整理報(bào)道,轉(zhuǎn)載請(qǐng)聯(lián)系本公眾號(hào)獲得授權(quán)。
1. star:9819|Weakly Supervised Disentanglement with Guarantees(弱監(jiān)督學(xué)習(xí))
論文:https://arxiv.org/pdf/1910.09772v2.pdf
代碼:https://github.com/google-research/google-research/tree/master/weak_disentangle
2. star:9819|Measuring Compositional Generalization: A Comprehensive Method on Realistic Data
論文:https://arxiv.org/pdf/1912.09713v1.pdf
代碼:https://github.com/google-research/google-research/tree/master/cfq
3. star:9819|Meta-Learning without Memorization(元學(xué)習(xí)/小樣本圖像分類)
論文:https://arxiv.org/pdf/1912.03820v3.pdf
代碼:https://github.com/google-research/google-research/tree/master/meta_learning_without_memorization
4. star:4977|U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation(圖像翻譯/無(wú)監(jiān)督)
論文:https://arxiv.org/pdf/1907.10830v4.pdf
代碼:https://github.com/taki0112/UGATIT

5. star:2106|On the Variance of the Adaptive Learning Rate and Beyond
論文:https://arxiv.org/pdf/1908.03265v3.pdf
代碼:https://github.com/LiyuanLucasLiu/RAdam

6. star:1469|DiffTaichi: Differentiable Programming for Physical Simulation
論文:https://arxiv.org/pdf/1910.00935v3.pdf
代碼:https://github.com/yuanming-hu/difftaichi
7. star:1018|Generative Models for Effective ML on Private, Decentralized Datasets
論文:https://arxiv.org/pdf/1911.06679v2.pdf
代碼:https://github.com/tensorflow/federated/tree/master/tensorflow_federated/python/research/gans

8. star:963|Behaviour Suite for Reinforcement Learning(強(qiáng)化學(xué)習(xí))
論文:https://arxiv.org/pdf/1908.03568v3.pdf
代碼:https://github.com/deepmind/bsuite
9. star:534|Contrastive Representation Distillation(知識(shí)蒸餾)
論文:https://arxiv.org/pdf/1910.10699v2.pdf
代碼:https://github.com/HobbitLong/RepDistiller

10. star:516|On the Relationship between Self-Attention and Convolutional Layers(注意力機(jī)制)
論文:https://arxiv.org/pdf/1911.03584v2.pdf
代碼:https://github.com/epfml/attention-cnn
11. star:469|AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
論文:https://arxiv.org/pdf/1912.02781v2.pdf
代碼:https://github.com/rwightman/pytorch-image-models

12. star:443|NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Search(神經(jīng)網(wǎng)絡(luò)架構(gòu)搜索)
論文:https://arxiv.org/pdf/2001.00326v2.pdf
代碼:https://github.com/D-X-Y/NAS-Projects
13. star:393|Once for All: Train One Network and Specialize it for Efficient Deployment(神經(jīng)網(wǎng)絡(luò)訓(xùn)練)
論文:https://openreview.net/pdf?id=HylxE1HKwS
代碼:https://github.com/mit-han-lab/once-for-all
14. star:246|BatchEnsemble: An Alternative Approach to Efficient Ensemble and Lifelong Learning(神經(jīng)網(wǎng)絡(luò)訓(xùn)練)
論文:https://arxiv.org/pdf/2002.06715v2.pdf
代碼:https://github.com/google/edward2
15. star:243|FasterSeg: Searching for Faster Real-time Semantic Segmentation(語(yǔ)義分割)
論文:https://arxiv.org/pdf/1912.10917v2.pdf
代碼:https://github.com/TAMU-VITA/FasterSeg

16. star:213|Contrastive Learning of Structured World Models
論文:https://arxiv.org/pdf/1911.12247v2.pdf
代碼:https://github.com/tkipf/c-swm

17. star:191|Real or Not Real, that is the Question(GAN)
論文:https://arxiv.org/pdf/2002.05512v1.pdf
代碼:https://github.com/kam1107/RealnessGAN
18. star:186|Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Driving(3D目標(biāo)檢測(cè))
論文:https://arxiv.org/pdf/1906.06310v3.pdf
代碼:https://github.com/mileyan/Pseudo_Lidar_V2
19. star:182|Learning to Explore using Active Neural SLAM(三維SLAM)
論文:https://arxiv.org/pdf/2004.05155v1.pdf
代碼:https://github.com/devendrachaplot/Neural-SLAM
20. star:175|Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification(行人重識(shí)別/無(wú)監(jiān)督)
論文:https://arxiv.org/pdf/2001.01526v2.pdf
代碼:https://github.com/yxgeee/MMT
21. star:132|AtomNAS: Fine-Grained End-to-End Neural Architecture Search(神經(jīng)網(wǎng)絡(luò)架構(gòu)搜索)
論文:https://arxiv.org/pdf/1912.09640v2.pdf
代碼:https://github.com/meijieru/AtomNAS
22. star:128|Strategies for Pre-training Graph Neural Networks(神經(jīng)網(wǎng)絡(luò)訓(xùn)練)
論文:https://arxiv.org/pdf/1905.12265v3.pdf
代碼:https://github.com/snap-stanford/pretrain-gnns/
23. star117|Towards Stabilizing Batch Statistics in Backward Propagation of Batch Normalization(歸一化)
論文:https://arxiv.org/pdf/2001.06838v2.pdf
代碼:https://github.com/megvii-model/MABN
24. star:107|DropEdge: Towards Deep Graph Convolutional Networks on Node Classification
論文:https://arxiv.org/pdf/1907.10903v4.pdf
代碼:https://github.com/DropEdge/DropEdge
25. star:107|Neural Arithmetic Units
論文:https://arxiv.org/pdf/2001.05016v1.pdf
代碼:https://github.com/AndreasMadsen/stable-nalu
26. star:106|Semantically-Guided Representation Learning for Self-Supervised Monocular Depth(單目深度估計(jì))
論文:https://arxiv.org/pdf/2002.12319v1.pdf
代碼:https://github.com/TRI-ML/packnet-sfm

27. star:100|Composition-based Multi-Relational Graph Convolutional Networks
論文:https://arxiv.org/pdf/1911.03082v2.pdf
代碼:https://github.com/malllabiisc/CompGCN
28. star:93|Deformable Kernels: Adapting Effective Receptive Fields for Object Deformation(圖像分割/目標(biāo)檢測(cè))
論文:https://arxiv.org/pdf/1910.02940v2.pdf
代碼:https://github.com/hangg7/deformable-kernels/

29. star:80|NAS evaluation is frustratingly hard(神經(jīng)網(wǎng)絡(luò)架構(gòu)搜索)
論文:https://arxiv.org/pdf/1912.12522v3.pdf
代碼:https://github.com/antoyang/NAS-Benchmark
30. star:74|Understanding and Robustifying Differentiable Architecture Search(圖像分類)
論文:https://arxiv.org/pdf/1909.09656v2.pdf
代碼:https://github.com/automl/RobustDARTS
31. star:72|Fast Neural Network Adaptation via Parameter Remapping and Architecture Search(圖像分類/目標(biāo)檢測(cè)/語(yǔ)義分割)
論文:https://arxiv.org/pdf/2001.02525v2.pdf
代碼:https://github.com/JaminFong/FNA
32. star:72|Capsules with Inverted Dot-Product Attention Routing(圖像分類)
論文:https://arxiv.org/pdf/2002.04764v2.pdf
代碼:https://github.com/apple/ml-capsules-inverted-attention-routing
33. star:53|Deep Semi-Supervised Anomaly Detection(異常檢測(cè))
論文:https://arxiv.org/pdf/1906.02694v2.pdf
代碼:https://arxiv.org/pdf/1906.02694v2.pdf

34. star:51|Network Deconvolution(圖像分類)
論文:https://arxiv.org/pdf/1905.11926v4.pdf
代碼:https://github.com/deconvolutionpaper/deconvolution
35. star:49|Pitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning(圖像分類)
論文:https://arxiv.org/pdf/2002.06470v1.pdf
代碼:https://github.com/bayesgroup/pytorch-ensembles
36. star:36|A Neural Dirichlet Process Mixture Model for Task-Free Continual Learning(圖像分類)
論文:https://arxiv.org/pdf/2001.00689v2.pdf
代碼:https://github.com/soochan-lee/CN-DPM
37. star:33|Empirical Bayes Transductive Meta-Learning with Synthetic Gradients(小樣本圖像分類/元學(xué)習(xí))
論文:https://openreview.net/pdf?id=Hkg-xgrYvH
代碼:https://github.com/hushell/sib_meta_learn
38. star:32|Query2box: Reasoning over Knowledge Graphs in Vector Space using Box Embeddings(知識(shí)圖譜)
論文:https://arxiv.org/pdf/2002.05969v2.pdf
代碼:https://github.com/hyren/query2box
39. star:27|Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps(圖像分類)
論文:https://openreview.net/pdf?id=BkgrBgSYDS
代碼:https://github.com/HazyResearch/learning-circuits
40. star:22|Fooling Detection Alone is Not Enough: Adversarial Attack against Multiple Object Tracking(多目標(biāo)跟蹤)
論文:https://openreview.net/pdf?id=rJl31TNYPr
代碼:https://github.com/anonymousjack/hijacking
打包下載:
鏈接:https://pan.baidu.com/s/1bM1ZAiX5FOY_65gBLAkoow?密碼:rjsn