2022 ICML 论文合集（附Code和数据集）

1. Sharp-MAML: Sharpness-Aware Model-Agnostic Meta Learning

Task：元学习

代码链接：https://github.com/mominabbass/sharp-maml

数据集：https://paperswithcode.com/dataset/mini-imagenet

2. Active Sampling for Min-Max Fairness

Task：算法公平性、回归

代码链接：https://github.com/amazon-research/active-sampling-for-minmax-fairness

数据集：None

3. Meaningfully debugging model mistakes using conceptual counterfactual explanations

代码链接：https://github.com/mertyg/debug-mistakes-cce

数据集：https://paperswithcode.com/dataset/imagenet

4. Hierarchical Shrinkage: Improving the accuracy and interpretability of tree-based models

代码链接：https://github.com/csinva/imodels

数据集：https://paperswithcode.com/dataset/friedman1

5. On the Convergence of the Shapley Value in Parametric Bayesian Learning Games

Task：贝叶斯推理

代码链接：https://github.com/XinyiYS/Parametric-Bayesian-Learning-Games

数据集：None

6. Individual Preference Stability for Clustering

Task：算法公平性

代码链接：https://github.com/amazon-research/ip-stability-for-clustering

数据集：None

7. Minimum Cost Intervention Design for Causal Effect Identification

代码链接：https://github.com/sinaakbarii/min_cost_intervention

数据集：None

8. How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models

Task：图像生成

代码链接：https://github.com/vanderschaarlab/evaluating-generative-models

数据集：None

9. Deploying Convolutional Networks on Untrusted Platforms Using 2D Holographic Reduced Representations

代码链接：https://github.com/neuromorphiccomputationresearchprogram/connectionist-symbolic-pseudo-secrets

数据集：None

10. Optimistic Linear Support and Successor Features as a Basis for Optimal Policy Transfer

Task：迁移学习

代码链接：https://github.com/lucasalegre/sfols

数据集：None

11. Structured Stochastic Gradient MCMC

Task：贝叶斯推理、变分推理

代码链接：https://github.com/ajboyd2/pytorch_lvi

数据集：https://paperswithcode.com/dataset/cifar-10

https://paperswithcode.com/dataset/svhn

https://paperswithcode.com/dataset/fashion-mnist

12. XAI for Transformers: Better Explanations through Conservative Propagation

代码链接：https://github.com/ameenali/xai_transformers

数据集：https://paperswithcode.com/dataset/sst

13. Neuro-Symbolic Language Modeling with Automaton-augmented Retrieval

Task：NLP-语言模型、检索

代码链接：https://github.com/neulab/knn-transformers

数据集：https://paperswithcode.com/dataset/wikitext-103

14. Minimax Classification under Concept Drift with Multidimensional Adaptation and Performance Guarantees

代码链接：https://github.com/machinelearningbcam/amrc-for-concept-drift-icml-2022

数据集：None

15. Scalable First-Order Bayesian Optimization via Structured Automatic Differentiation

代码链接：https://github.com/sebastianament/covariancefunctions.jl

数据集：None

16. Towards Understanding Sharpness-Aware Minimization

代码链接：https://github.com/tml-epfl/understanding-sam

数据集：https://paperswithcode.com/dataset/cifar-100

17. Image-to-Image Regression with Distribution-Free Uncertainty Quantification and Applications in Imaging

Task：超分辨率、回归

代码链接：https://github.com/aangelopoulos/im2im-uq

数据集：https://paperswithcode.com/dataset/fastmri

18. Online Balanced Experimental Design

代码链接：https://github.com/ddimmery/balancer-package

数据集：None

19. Thresholded Lasso Bandit

代码链接：https://github.com/cyberagentailab/thresholded-lasso-bandit

数据集：None

20. From Noisy Prediction to True Label: Noisy Prediction Calibration via Generative Model

Task：Learning with noisy labels

代码链接：https://github.com/BaeHeeSun/NPC

数据集：https://paperswithcode.com/dataset/clothing1m

21. data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language

Task：自监督学习、语音识别、自然语言理解、自然语言推理、图像分类

代码链接：https://github.com/pytorch/fairseq/tree/main/examples/data2vec

数据集：https://paperswithcode.com/dataset/libri-light

22. Gaussian Mixture Variational Autoencoder with Contrastive Learning for Multi-Label Classification

Task：对比学习、多标签分类

代码链接：https://github.com/junwenbai/c-gmvae

数据集：None

23. A3T: Alignment-Aware Acoustic and Text Pretraining for Speech Synthesis and Editing

Task：表示学习、语音识别、语音合成、Speech-to-Text Translation

代码链接：https://github.com/richardbaihe/a3t

数据集：https://paperswithcode.com/dataset/vctk

24. Certified Neural Network Watermarks with Randomized Smoothing

代码链接：https://github.com/arpitbansal297/certified_watermarks

数据集：None

25. Learning Stable Classifiers by Transferring Unstable Features

Task：迁移学习

代码链接：https://github.com/YujiaBao/Tofu

数据集：https://paperswithcode.com/dataset/celeba

26. Estimating the Optimal Covariance with Imperfect Mean in Diffusion Probabilistic Models

代码链接：https://github.com/baofff/Extended-Analytic-DPM

数据集：https://paperswithcode.com/dataset/lsun

27. On the Surrogate Gap between Contrastive and Supervised Losses

Task：表示学习、数据增强

代码链接：https://github.com/nzw0301/gap-contrastive-and-supervised-losses

数据集：https://paperswithcode.com/dataset/cifar-100

28. Representation Topology Divergence: A Method for Comparing Neural Network Representations.

Task：集成学习、迁移学习、Disentanglement

代码链接：https://github.com/ilyatrofimov/rtd

数据集：https://paperswithcode.com/dataset/cifar-100

29. Imitation Learning by Estimating Expertise of Demonstrators

Task：模仿学习

代码链接：https://github.com/stanford-iliad/ileed

数据集：None

30. Neural Inverse Kinematic

代码链接：https://github.com/RaphaelBensTAU/NeuralInverseKinematics

数据集：None

31. Volatility Based Kernels and Moving Average Means for Accurate Forecasting with Gaussian Processes

Task：高斯过程

代码链接：https://github.com/g-benton/volt

数据集：None

32. Gradient Descent on Neurons and its Link to Approximate Second-order Optimization

代码链接：https://github.com/freedbee/neuron_descent_and_kfac

数据集：https://paperswithcode.com/dataset/fashion-mnist

33. Skin Deep Unlearning: Artefact and Instrument Debiasing in the Context of Melanoma Classification

Task：医学图像分类

代码链接：https://github.com/pbevan1/Skin-Deep-Unlearning

数据集：https://paperswithcode.com/dataset/isic-2020-challenge-dataset

34. Approximate Bayesian Computation with Domain Expert in the Loop

Task：主动学习、数据降维

代码链接：https://github.com/lfilstro/hitl-abc

数据集：https://paperswithcode.com/dataset/isic-2020-challenge-dataset

35. Personalization Improves Privacy-Accuracy Tradeoffs in Federated Learning

Task：联邦学习、个性化联邦学习、随机优化

代码链接：https://github.com/albietz/ppsgd

数据集：None

36. Non-Vacuous Generalisation Bounds for Shallow Neural Networks

代码链接：https://github.com/biggs/shallow-nets

数据集：https://paperswithcode.com/dataset/fashion-mnist

37. Scalable Spike-and-Slab

Task：变量选择

代码链接：https://github.com/niloyb/scalespikeslab

数据集：None

38. Breaking Down Out-of-Distribution Detection: Many Methods Based on OOD Training Data Estimate a Combination of the Same Core Quantities

Task：OOD检测、分布外检测

代码链接：https://github.com/j-cb/breaking_down_ood_detection

数据集：https://paperswithcode.com/dataset/celeba

39. Optimizing Sequential Experimental Design with Deep Reinforcement Learning

Task：强化学习

代码链接：https://github.com/csiro-mlai/RL-BOED

数据集：None

40. Generalized Results for the Existence and Consistency of the MLE in the Bradley-Terry-Luce Model

代码链接：https://github.com/heejongbong/mmpc

数据集：None

41. How to Train Your Wide Neural Network Without Backprop: An Input-Weight Alignment Perspective

代码链接：https://github.com/fietelab/wide-network-alignment

数据集：https://paperswithcode.com/dataset/cifar-10

42. Improving Language Models by Retrieving from Trillions of Tokens

Task：语言模型、QA、检索

代码链接：https://github.com/labmlai/annotated_deep_learning_paper_implementations

数据集：https://paperswithcode.com/dataset/natural-questions

43. Lie Point Symmetry Data Augmentation for Neural PDE Solvers

Task：数据增强

代码链接：https://github.com/brandstetter-johannes/lpsda

数据集：None

44. An iterative clustering algorithm for the Contextual Stochastic Block Model with optimality guarantees

Task：随机块模型

代码链接：https://github.com/glmbraun/csbm

数据集：None

45. Tractable Dendritic RNNs for Reconstructing Nonlinear Dynamical Systems

Task：时间序列、变分推理

代码链接：https://github.com/durstewitzlab/dendplrnn

数据集：None

46. Learning to Predict Graphs with Fused Gromov-Wasserstein Barycenters

Task：回归

代码链接：https://github.com/lmotte/graph-prediction-with-fused-gromov-wasserstein

数据集：None

47. IGLUE: A Benchmark for Transfer Learning across Modalities, Tasks, and Languages

Task：跨模态检索、小样本学习、迁移学习、QA

代码链接：https://github.com/e-bug/volta

数据集：https://paperswithcode.com/dataset/iglue

48. Measuring dissimilarity with diffeomorphism invariance

代码链接：https://github.com/theophilec/diffy

数据集：None

49. Gaussian Process Uniform Error Bounds with Unknown Hyperparameters for Safety-Critical Applications

Task：高斯过程

代码链接：https://github.com/aCapone1/gauss_proc_unknown_hyp

数据集：None

50. Burst-Dependent Plasticity and Dendritic Amplification Support Target-Based Learning and Hierarchical Imitation Learning

Task：模仿学习

代码链接：https://github.com/cristianocapone/lttb

数据集：None

51. RECAPP: Crafting a More Efficient Catalyst for Convex Optimization

代码链接：https://github.com/yaircarmon/recapp

数据集：None

52. YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for Everyone

Task：零样本学习、语音合成

代码链接：https://github.com/coqui-ai/TTS

数据集：https://paperswithcode.com/dataset/voxceleb1

53. Stabilizing Off-Policy Deep Reinforcement Learning from Pixels

Task：数据增强、强化学习

代码链接：https://github.com/aladoro/stabilizing-off-policy-rl

数据集：https://paperswithcode.com/dataset/deepmind-control-suite

54. Robust Imitation Learning against Variations in Environment Dynamics

Task：模仿学习

代码链接：https://github.com/jongseongchae/rime

数据集：https://paperswithcode.com/dataset/mujoco

55. Revisiting Label Smoothing and Knowledge Distillation Compatibility: What was Missing?

Task：图像分类、知识蒸馏、机器翻译

代码链接：https://github.com/sutd-visual-computing-group/LS-KD-compatibility

数据集：https://paperswithcode.com/dataset/imagenet

56. Learning Bellman Complete Representations for Offline Policy Evaluation

Task：表示学习

代码链接：https://github.com/causalml/bcrl

数据集：https://paperswithcode.com/dataset/deepmind-control-suite

57. Sample Efficient Learning of Predictors that Complement Humans

Task：主动学习

代码链接：https://github.com/clinicalml/active_learn_to_defer

数据集：https://paperswithcode.com/dataset/cifar-10

58. Coarsening the Granularity: Towards Structurally Sparse Lottery Tickets

代码链接：https://github.com/vita-group/structure-lth

数据集：https://paperswithcode.com/dataset/tiny-imagenet

59. Learning Domain Adaptive Object Detection with Probabilistic Teacher

Task：目标检测

代码链接：https://github.com/hikvision-research/probabilisticteacher

数据集：https://paperswithcode.com/dataset/cityscapes

60. Perfectly Balanced: Improving Transfer and Robustness of Supervised Contrastive Learning

Task：对比学习、数据增强

代码链接：https://github.com/HazyResearch/thanos-code

数据集：https://paperswithcode.com/dataset/mnist

61. On Collective Robustness of Bagging Against Data Poisoning

代码链接：https://github.com/emiyalzn/icml22-crb

数据集：https://cs.paperswithcode.com/dataset/cifar-10

62. Structure-Aware Transformer for Graph Representation Learning

Task：图表示学习、表示学习

代码链接：https://github.com/borgwardtlab/sat

数据集：https://paperswithcode.com/dataset/ogb

63. Optimization-Induced Graph Implicit Nonlinear Diffusion

代码链接：https://github.com/7qchen/gind

数据集：https://paperswithcode.com/dataset/ppi

64. Robust Meta-learning with Sampling Noise and Label Noise via Eigen-Reptile

Task：小样本学习

代码链接：https://github.com/anfeather/eigen-reptile

数据集：https://paperswithcode.com/dataset/mini-imagenet

65. Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

Task：迁移学习

代码链接：https://github.com/vita-group/double-win-lth

数据集：https://paperswithcode.com/dataset/svhn

66. Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness

代码链接：https://github.com/vita-group/linearity-grafting

数据集：https://paperswithcode.com/dataset/svhn

67. Task-aware Privacy Preservation for Multi-dimensional Data

代码链接：https://github.com/chengjiangnan/task_aware_privacy

数据集：None

68. Adversarially Trained Actor Critic for Offline Reinforcement Learning

Task：强化学习、离线强化学习

代码链接：https://github.com/microsoft/atac

数据集：https://paperswithcode.com/dataset/d4rl

69. RieszNet and ForestRiesz: Automatic Debiased Machine Learning with Neural Nets and Random Forests

Task：BIG-bench 机器学习

代码链接：https://github.com/victor5as/rieszlearning

数据集：None

70. Selective Network Linearization for Efficient Private Inference

代码链接：https://github.com/nyu-dice-lab/selective_network_linearization

数据集：None

71. From block-Toeplitz matrices to differential equations on graphs: towards a general theory for scalable masked Transformers

Task：图注意力

代码链接：https://github.com/hl-hanlin/gkat

数据集：https://paperswithcode.com/dataset/pubmed

72. TPC: Transformation-Specific Smoothing for Point Cloud Models

Task：自动驾驶

代码链接：https://github.com/qianhewu/point-cloud-smoothing

数据集：https://paperswithcode.com/dataset/modelnet

73. Context-Aware Drift Detection

代码链接：https://github.com/SeldonIO/alibi-detect

数据集：https://paperswithcode.com/dataset/imagenet

74. Diffusion bridges vector quantized variational autoencoders

代码链接：https://github.com/maxjcohen/diffusion-bridges

数据集：https://paperswithcode.com/dataset/mini-imagenet

75. Label-Free Explainability for Unsupervised Models

Task：特征重要性

代码链接：https://github.com/vanderschaarlab/mlforhealthlabpub

数据集：https://paperswithcode.com/dataset/dsprites

76. Evaluating the Adversarial Robustness of Adaptive Test-time Defenses

Task：对抗鲁棒性、图像分类

代码链接：https://github.com/fra31/evaluating-adaptive-test-time-defenses

数据集：https://paperswithcode.com/dataset/robustbench

77. Continuous Control with Action Quantization from Demonstrations

Task：模仿学习、模型量化

代码链接：https://github.com/google-research/google-research

数据集：None

78. Dialog Inpainting: Turning Documents into Dialogs

Task：QA、对话式问答、检索

代码链接：https://github.com/google-research/dialog-inpainting

数据集：https://paperswithcode.com/dataset/ms-marco

79. DisPFL: Towards Communication-Efficient Personalized Federated Learning via Decentralized Sparse Training

Task：个性化联邦学习

代码链接：https://github.com/rong-dai/dispfl

数据集：https://paperswithcode.com/dataset/cifar-100

80. Unsupervised Image Representation Learning with Deep Latent Particles

Task：视频预测、表示学习、无监督人脸关键点检测

代码链接：https://github.com/taldatech/deep-latent-particles-pytorch

数据集：https://paperswithcode.com/dataset/celeba

81. Monarch: Expressive Structured Matrices for Efficient and Accurate Training

Task：语言模型

代码链接：https://github.com/hazyresearch/monarch

数据集：https://paperswithcode.com/dataset/wikitext-2

82. Test-Time Training Can Close the Natural Distribution Shift Performance Gap in Deep Learning Based Compressed Sensing

Task：影像重建、压缩传感、领域适应

代码链接：https://github.com/mli-lab/ttt_for_deep_learning_cs

数据集：https://paperswithcode.com/dataset/fastmri

83. Knowledge Base Question Answering by Case-based Reasoning over Subgraphs

Task：QA、知识库问答

代码链接：https://github.com/rajarshd/cbr-subg

数据集：https://paperswithcode.com/dataset/metaqa

84. Distinguishing rule and exemplar-based generalization in learning systems

Task：BIG-bench 机器学习、数据增强、算法公平性

代码链接：https://github.com/eringrant/icml-2022-rules-vs-exemplars

数据集：https://paperswithcode.com/dataset/celeba

85. Robust Multi-Objective Bayesian Optimization Under Input Noise

代码链接：https://github.com/facebookresearch/robust_mobo

数据集：None

86. Attentional Meta-learners for Few-shot Polythetic Classification

Task：小样本学习

代码链接：https://github.com/rvinas/polythetic_metalearning

数据集：https://paperswithcode.com/dataset/tieredimagenet

87. Born-Infeld (BI) for AI: Energy-Conserving Descent (ECD) for Optimization

代码链接：https://github.com/gbdl/bbi

数据集：https://paperswithcode.com/dataset/mnist

88. Error-driven Input Modulation: Solving the Credit Assignment Problem without a Backward Pass

代码链接：https://github.com/giorgiad/pepita

数据集：None

89. DreamerPro: Reconstruction-Free Model-Based Reinforcement Learning with Prototypical Representations

Task：强化学习、有模型强化学习

代码链接：https://github.com/fdeng18/dreamer-pro

数据集：https://paperswithcode.com/dataset/deepmind-control-suite

90. NeuralEF: Deconstructing Kernels by Deep Neural Networks

Task：图像分类

代码链接：https://github.com/thudzj/neuraleigenfunction

数据集：https://paperswithcode.com/dataset/cifar-10

91. Generalization and Robustness Implications in Object-Centric Learning

Task：表示学习、感应偏置、系统概括

代码链接：https://github.com/addtt/object-centric-library

数据集：https://paperswithcode.com/dataset/clevr

92. Fair Generalized Linear Models with a Convex Penalty

Task：算法公平性

代码链接：https://github.com/hyungrok-do/fair-glm-cvx

数据集：None

93. On the Adversarial Robustness of Causal Algorithmic Recourse

Task：对抗鲁棒性

代码链接：https://github.com/ricardodominguez/adversariallyrobustrecourse

数据集：None

94. Finding the Task-Optimal Low-Bit Sub-Distribution in Deep Neural Networks

Task：目标检测、模型压缩、模型量化、图像分类

代码链接：https://github.com/RunpeiDong/DGMS

数据集：https://paperswithcode.com/dataset/cub-200-2011

95. PACE: A Parallelizable Computation Encoder for Directed Acyclic Graphs

Task：神经架构搜索

代码链接：https://github.com/zehao-dong/pace

数据集：https://paperswithcode.com/dataset/ogb

96. Privacy for Free: How does Dataset Condensation Help Privacy?

Task：数据集压缩、隐私保护

代码链接：https://github.com/Guang000/Awesome-Dataset-Distillation

数据集：https://paperswithcode.com/dataset/mnist

97. Fast rates for noisy interpolation require rethinking the effect of inductive bias

Task：感应偏置

代码链接：https://github.com/nickruggeri/fast-rates-for-noisy-interpolation

数据集：None

98. Adapting to Mixing Time in Stochastic Optimization with Markovian Data

Task：随机优化

代码链接：https://github.com/Rondorf/BOReL

数据集：None

99. TACTiS: Transformer-Attentional Copulas for Time Series

Task：时间序列预测

代码链接：https://github.com/ServiceNow/tactis

数据集：None

100. Learning Iterative Reasoning through Energy Minimization

Task：图像分类、目标识别

代码链接：https://github.com/yilundu/irem_code_release

数据集：None

101. A Context-Integrated Transformer-Based Neural Network for Auction Design

代码链接：https://github.com/zjduan/CITransNet

数据集：None

102. From data to functa: Your data point is a function and you can treat it like one

Task：缺失值处理、新视图生成

代码链接：https://github.com/deepmind/functa

数据集：https://paperswithcode.com/dataset/shapenet

103. LIMO: Latent Inceptionism for Targeted Molecule Generation

Task：药物发现、高斯过程

代码链接：https://github.com/rose-stl-lab/limo

数据集：https://paperswithcode.com/dataset/zinc

104. FedNew: A Communication-Efficient and Privacy-Preserving Newton-Type Method for Federated Learning

Task：联邦学习、隐私保护、模型量化

代码链接：https://github.com/aelgabli/fednew

数据集：None

105. For Learning in Symmetric Teams, Local Optima are Global Nash Equilibria

代码链接：https://github.com/scottemmons/coordination

数据集：None

106. Towards Scaling Difference Target Propagation by Learning Backprop Targets

代码链接：https://github.com/bptargetdtp/scalabledtp

数据集：None

107. Understanding Dataset Difficulty with V-Usable Information

代码链接：https://github.com/kawine/dataset_difficulty

数据集：https://paperswithcode.com/dataset/snli

108. Head2Toe: Utilizing Intermediate Representations for Better Transfer Learning

Task：迁移学习

代码链接：https://github.com/google-research/head2toe

数据集：https://paperswithcode.com/dataset/oxford-102-flower

109. Variational Sparse Coding with Learned Thresholding

Task：变分推理

代码链接：https://github.com/kfallah/variational-sparse-coding

数据集：https://paperswithcode.com/dataset/fashion-mnist

110. Training Discrete Deep Generative Models via Gapped Straight-Through Estimator

Task：ListOps、强化学习

代码链接：https://github.com/chijames/gst

数据集：https://paperswithcode.com/dataset/listops

111. DRIBO: Robust Deep Reinforcement Learning via Multi-View Information Bottleneck

Task：强化学习、表示学习

代码链接：https://github.com/BU-DEPEND-Lab/DRIBO

数据集：https://paperswithcode.com/dataset/kinetics

112. Variational Wasserstein gradient flow

代码链接：https://github.com/sbyebss/variational_wgf

数据集：https://paperswithcode.com/dataset/cifar-10

113. Data Determines Distributional Robustness in Contrastive Language Image Pre-training (CLIP)

Task：图像分类

代码链接：https://github.com/mlfoundations/imagenet-captions

数据集：https://paperswithcode.com/dataset/yfcc100m

114. An Equivalence Between Data Poisoning and Byzantine Gradient Attacks

Task：Data Poisoning、联邦学习、个性化联邦学习

代码链接：https://github.com/lpd-epfl/attack_equivalence

数据集：None

115. Investigating Generalization by Controlling Normalized Margin

Task：学习理论

代码链接：https://github.com/alexfarhang/margin

数据集：None

116. Private frequency estimation via projective geometry

代码链接：https://github.com/minilek/private_frequency_oracles

数据集：None

117. Coordinated Double Machine Learning

Task：BIG-bench 机器学习

代码链接：https://github.com/nitaifingerhut/c-dml

数据集：None

118. Conformal Prediction Sets with Limited False Positives

代码链接：https://github.com/ajfisch/conformal-fp

数据集：None

119. Contrastive Mixture of Posteriors for Counterfactual Inference, Data Integration and Fairness

Task：反事实推理、数据集成、算法公平性、表示学习

代码链接：https://github.com/benevolentai/comp

数据集：None

120. A Neural Tangent Kernel Perspective of GANs

代码链接：https://github.com/google/neural-tangents

数据集：None

121. SPDY: Accurate Pruning with Speedup Guarantees

Task：模型压缩

代码链接：https://github.com/ist-daslab/spdy

数据集：https://paperswithcode.com/dataset/squad

122. Scaling Structured Inference with Randomization

代码链接：https://github.com/franxyao/rdp

数据集：None

123. DepthShrinker: A New Compression Paradigm Towards Boosting Real-Hardware Efficiency of Compact Neural Networks

代码链接：https://github.com/rice-eic/depthshrinker

数据集：None

124. p-Laplacian Based Graph Neural Networks

Task：节点分类

代码链接：https://github.com/guoji-fu/pgnns

数据集：https://paperswithcode.com/dataset/wiki-squirrel

125. Deep Reference Priors: What is the best way to pretrain a model?

Task：半监督图像分类、迁移学习

代码链接：https://github.com/rahul13ramesh/deep_reference_priors

数据集：https://paperswithcode.com/dataset/cifar-100

126. Rethinking Image-Scaling Attacks: The Interplay Between Vulnerabilities in Machine Learning Systems

Task：BIG-bench 机器学习

代码链接：https://github.com/wi-pi/rethinking-image-scaling-attacks

数据集：https://paperswithcode.com/dataset/celeba

127. Fast and Reliable Evaluation of Adversarial Robustness with Minimum-Margin Attack

Task：对抗鲁棒性

代码链接：https://github.com/sjtubrian/mm-attack

数据集：None

128. Value Function based Difference-of-Convex Algorithm for Bilevel Hyperparameter Selection Problems

代码链接：https://github.com/sustech-optimization/vf-idca

数据集：None

129. Learning to Incorporate Texture Saliency Adaptive Attention to Image Cartoonization

Task：Image-to-Image Translation、风格迁移、Unsupervised Image-To-Image Translation

代码链接：https://github.com/xianggao1102/learning-to-incorporate-texture-saliency-adaptive-attention-to-image-cartoonization

数据集：None

130. Stochastic smoothing of the top-K calibrated hinge loss for deep imbalanced classification

Task：不均衡分类

代码链接：https://github.com/garcinc/noised-topk

数据集：https://paperswithcode.com/dataset/imagenet-lt

131. A Functional Information Perspective on Model Interpretation

代码链接：https://github.com/nitaytech/functionalexplanation

数据集：https://paperswithcode.com/dataset/imdb-movie-reviews

132. Inducing Causal Structure for Interpretable Neural Networks

Task：数据增强、自然语言推理

代码链接：https://github.com/frankaging/interchange-intervention-training

数据集：https://paperswithcode.com/dataset/scan

133. Near-Exact Recovery for Tomographic Inverse Problems via Deep Learning

Task：Computed Tomography

代码链接：https://github.com/jmaces/aapm-ct-challenge

数据集：None

134. Equivariance versus Augmentation for Spherical Images

Task：数据增强、图像分类、语义分割

代码链接：https://github.com/janegerken/sem_seg_s2cnn

数据集：https://paperswithcode.com/dataset/fashion-mnist

135. Plug-In Inversion: Model-Agnostic Inversion for Vision with Data Augmentations

Task：图像分类

代码链接：https://github.com/youranonymousefriend/plugininversion

数据集：https://paperswithcode.com/dataset/imagenet

136. SCHA-VAE: Hierarchical Context Aggregation for Few-Shot Generation

Task：小样本学习、分布外泛化

代码链接：https://github.com/georgosgeorgos/hierarchical-few-shot-generative-models

数据集：https://paperswithcode.com/dataset/celeba

137. It’s Raw! Audio Generation with State-Space Models

Task：音频生成、密度估计、音乐合成

代码链接：https://github.com/hazyresearch/state-spaces

数据集：None

138. RankSim: Ranking Similarity Regularization for Deep Imbalanced Regression

Task：归纳偏置、回归

代码链接：https://github.com/BorealisAI/ranksim-imbalanced-regression

数据集：None

139. Secure Distributed Training at Scale

Task：分布式优化、图像分类、语言模型

代码链接：https://github.com/yandex-research/btard

数据集：None

140. The State of Sparse Training in Deep Reinforcement Learning

Task：强化学习

代码链接：https://github.com/google-research/rigl/tree/master/rigl/rl

数据集：https://paperswithcode.com/dataset/mujoco

141. Causal Inference Through the Structural Causal Marginal Problem

Task：因果推断、反事实推理

代码链接：https://github.com/lgresele/structural-causal-marginal

数据集：None

142. Mirror Learning: A Unifying Framework of Policy Optimisation

代码链接：https://github.com/znowu/mirror-learning

数据集：None

143. Variational Mixtures of ODEs for Inferring Cellular Gene Expression Dynamics

代码链接：https://github.com/welch-lab/velovae

数据集：None

144. Leveraging Approximate Symbolic Models for Reinforcement Learning via Skill Diversity

Task：强化学习

代码链接：https://github.com/GuanSuns/ASGRL

数据集：None

145. Bounding Training Data Reconstruction in Private (Deep) Learning

代码链接：https://github.com/facebookresearch/bounding_data_reconstruction

数据集：https://paperswithcode.com/dataset/cifar-10

146. NISPA: Neuro-Inspired Stability-Plasticity Adaptation for Continual Learning in Sparse Networks

Task：持续学习

代码链接：https://github.com/burakgurbuz97/nispa

数据集：https://paperswithcode.com/dataset/emnist

147. Active Learning on a Budget: Opposite Strategies Suit High and Low Budgets

Task：主动学习

代码链接：https://github.com/avihu111/typiclust

数据集：https://paperswithcode.com/dataset/cifar-100

148. You Only Cut Once: Boosting Data Augmentation with a Single Cut

Task：数据增强

代码链接：https://github.com/junlinhan/yoco

数据集：https://paperswithcode.com/dataset/imagenet-a

149. Scalable MCMC Sampling for Nonsymmetric Determinantal Point Processes

Task：点过程

代码链接：https://github.com/insuhan/ndpp-mcmc-sampling

数据集：None

150. Adversarial Attacks on Gaussian Process Bandits

Task：对抗性攻击、高斯过程

代码链接：https://github.com/eric-vader/attack-bo

数据集：None