Skip to content

Latest commit

 

History

History
1377 lines (824 loc) · 69.6 KB

ICLR2021-Papers-with-Code.md

File metadata and controls

1377 lines (824 loc) · 69.6 KB
Raw

ICLR 2021 论文和开源项目合集

本仓库旨在收集ICLR最新研究进展，尤其是LLM方面，涉及NLP领域的各个方向，此项目长期不定时更新。

欢迎watch和fork！不过给个star⭐就更好了❤️。

知乎地址：ShuYini

微信公众号: AINLPer（每日更新，欢迎关注）

ICLR2021 Accept Paper With Code

1、Robust Reinforcement Learning on State Observations with Learned Optimal Adversary

2、Modeling the Second Player in Distributionally Robust Optimization

3、DEBERTA: DECODING-ENHANCED BERT WITH DISENTANGLED ATTENTION

4、Sparse encoding for more-interpretable feature-selecting representations in probabilistic matrix factorization

5、Neural networks with late-phase weights

6、ResNet After All: Neural ODEs and Their Numerical Solution

7、Transformer protein language models are unsupervised structure learners

8、Mapping the Timescale Organization of Neural Language Models

9、Differentiable Trust Region Layers for Deep Reinforcement Learning

10、Multiscale Score Matching for Out-of-Distribution Detection

11、Learning Associative Inference Using Fast Weight Memory

12、Hopfield Networks is All You Need

13、Learning to Set Waypoints for Audio-Visual Navigation

14、Variational Information Bottleneck for Effective Low-Resource Fine-Tuning

15、Seq2Tens: An Efficient Representation of Sequences by Low-Rank Tensor Projections

16、Model-based micro-data reinforcement learning: what are the crucial model properties and which model to choose?

17、Colorization Transformer

18、Is Attention Better Than Matrix Decomposition?

19、Clairvoyance: A Pipeline Toolkit for Medical Time Series

20、GraPPa: Grammar-Augmented Pre-Training for Table Semantic Parsing

21、Characterizing signal propagation to close the performance gap in unnormalized ResNets

22、Prototypical Contrastive Learning of Unsupervised Representations

23、Boost then Convolve: Gradient Boosting Meets Graph Neural Networks

24、On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines

25、End-to-End Egospheric Spatial Memory

26、Linear Mode Connectivity in Multitask and Continual Learning

27、Disentangling 3D Prototypical Networks for Few-Shot Concept Learning

28、Learning Energy-Based Models by Diffusion Recovery Likelihood

29、Training GANs with Stronger Augmentations via Contrastive Discriminator

30、On Graph Neural Networks versus Graph-Augmented MLPs

31、Adaptive Procedural Task Generation for Hard-Exploration Problems

32、DialoGraph: Incorporating Interpretable Strategy-Graph Networks into Negotiation Dialogues

33、Solving Compositional Reinforcement Learning Problems via Task Reduction

34、Discovering Diverse Multi-Agent Strategic Behavior via Reward Randomization

35、WaNet - Imperceptible Warping-based Backdoor Attack

36、Contextual Transformation Networks for Online Continual Learning

37、Reset-Free Lifelong Learning with Skill-Space Planning

38、Revisiting Locally Supervised Learning: an Alternative to End-to-end Training

39、CaPC Learning: Confidential and Private Collaborative Learning

40、Efficient Empowerment Estimation for Unsupervised Stabilization

41、SSD: A Unified Framework for Self-Supervised Outlier Detection

42、Robust Overfitting may be mitigated by properly learned smoothening

43、Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks

44、Property Controllable Variational Autoencoder via Invertible Mutual Dependence

45、Learning Task Decomposition with Ordered Memory Policy Network

46、Conditional Generative Modeling via Learning the Latent Space

47、Grounding Physical Concepts of Objects and Events Through Dynamic Visual Reasoning

48、gradSim: Differentiable simulation for system identification and visuomotor control

49、 Decentralized Attribution of Generative Models

50、Extreme Memorization via Scale of Initialization

51、GANs Can Play Lottery Tickets Too

52、Learning to Deceive Knowledge Graph Augmented Models via Targeted Perturbation

53、Into the Wild with AudioScope: Unsupervised Audio-Visual Separation of On-Screen Sounds

54、Layer-adaptive Sparsity for the Magnitude-based Pruning

55、Understanding and Improving Encoder Layer Fusion in Sequence-to-Sequence Learning

56、CompOFA – Compound Once-For-All Networks for Faster Multi-Platform Deployment

57、Are wider nets better given the same number of parameters?

58、SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularization

59、PolarNet: Learning to Optimize Polar Keypoints for Keypoint Based Object Detection

60、Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding

61、Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective

62、Contrastive Syn-to-Real Generalization

63、Pre-training Text-to-Text Transformers for Concept-centric Common Sense

64、Benchmarks for Deep Off-Policy Evaluation

65、Learning Long-term Visual Dynamics with Region Proposal Interaction Networks

66、IsarStep: a Benchmark for High-level Mathematical Reasoning

67、Simple Spectral Graph Convolution

68、Beyond Categorical Label Representations for Image Classification

69、Towards Robust Neural Networks via Close-loop Control

70、On Fast Adversarial Robustness Adaptation in Model-Agnostic Meta-Learning

71、Understanding the failure modes of out-of-distribution generalization

72、Reinforcement Learning with Random Delays

73、Faster Binary Embeddings for Preserving Euclidean Distances

74、Scalable Transfer Learning with Expert Models

75、Signatory: differentiable computations of the signature and logsignature transforms, on both CPU and GPU

76、Deep Partition Aggregation: Provable Defenses against General Poisoning Attacks

77、C-Learning: Horizon-Aware Cumulative Accessibility Estimation

78、Decoupling Global and Local Representations via Invertible Generative Flows

79、Viewmaker Networks: Learning Views for Unsupervised Representation Learning

80、Long Live the Lottery: The Existence of Winning Tickets in Lifelong Learning

81、Neural representation and generation for RNA secondary structures

82、Learning A Minimax Optimizer: A Pilot Study

83、Fuzzy Tiling Activations: A Simple Approach to Learning Sparse Representations Online

84、Calibration tests beyond classification

85、Class Normalization for (Continual)? Generalized Zero-Shot Learning

86、UMEC: Unified model and embedding compression for efficient recommendation systems

87、No MCMC for me: Amortized sampling for fast and stable training of energy-based models

88、Learning and Evaluating Representations for Deep One-Class Classification

89、InfoBERT: Improving Robustness of Language Models from An Information Theoretic Perspective

90、CopulaGNN: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks

91、Using latent space regression to analyze and leverage compositionality in GANs

92、Zero-Cost Proxies for Lightweight NAS

93、IOT: Instance-wise Layer Reordering for Transformer Structures

94、BREEDS: Benchmarks for Subpopulation Shift

95、Federated Learning via Posterior Averaging: A New Perspective and Practical Algorithms

96、On the Impossibility of Global Convergence in Multi-Loss Optimization

97、Spatial Dependency Networks: Neural Layers for Improved Generative Image Modeling

98、Interpretable Neural Architecture Search via Bayesian Optimisation with Weisfeiler-Lehman Kernels

99、ChipNet: Budget-Aware Pruning with Heaviside Continuous Approximations

100、AdaSpeech: Adaptive Text to Speech for Custom Voice

101、Learning Safe Multi-agent Control with Decentralized Neural Barrier Certificates

102、Progressive Skeletonization: Trimming more fat from a network at initialization

103、Estimating and Evaluating Regression Predictive Uncertainty in Deep Object Detectors

104、Learnable Embedding sizes for Recommender Systems

105、Learning Neural Generative Dynamics for Molecular Conformation Generation

106、Ask Your Humans: Using Human Instructions to Improve Generalization in Reinforcement Learning

107、FastSpeech 2: Fast and High-Quality End-to-End Text to Speech

108、Revisiting Hierarchical Approach for Persistent Long-Term Video Prediction

109、Taming GANs with Lookahead-Minmax

110、QPLEX: Duplex Dueling Multi-Agent Q-Learning

111、Acting in Delayed Environments with Non-Stationary Markov Policies

112、Prototypical Representation Learning for Relation Extraction

113、Deployment-Efficient Reinforcement Learning via Model-Based Offline Optimization

114、Efficient Conformal Prediction via Cascaded Inference with Expanded Admission

115、Uncertainty Estimation and Calibration with Finite-State Probabilistic RNNs

116、Meta-learning Symmetries by Reparameterization

117、When Optimizing $f$-Divergence is Robust with Label Noise

118、Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval

119、What Makes Instance Discrimination Good for Transfer Learning?

120、MODALS: Modality-agnostic Automated Data Augmentation in the Latent Space

121、ALFWorld: Aligning Text and Embodied Environments for Interactive Learning

122、Latent Skill Planning for Exploration and Transfer

123、Combining Ensembles and Data Augmentation Can Harm Your Calibration

124、AdaGCN: Adaboosting Graph Convolutional Networks into Deep Models

125、CoCo: Controllable Counterfactuals for Evaluating Dialogue State Trackers

126、Loss Function Discovery for Object Detection via Convergence-Simulation Driven Search

127、Generating Adversarial Computer Programs using Optimized Obfuscations

128、Anchor & Transform: Learning Sparse Embeddings for Large Vocabularies

129、Fast convergence of stochastic subgradient method under interpolation

130、Task-Agnostic Morphology Evolution

131、WaveGrad: Estimating Gradients for Waveform Generation

132、C-Learning: Learning to Achieve Goals via Recursive Classification

133、Revisiting Few-sample BERT Fine-tuning

134、Incorporating Symmetry into Deep Dynamics Models for Improved Generalization

135、Contrastive Learning with Hard Negative Samples

136、Scaling Symbolic Methods using Gradients for Neural Model Explanation

137、Risk-Averse Offline Reinforcement Learning

138、Fair Mixup: Fairness via Interpolation

139、Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis

140、Learning with Instance-Dependent Label Noise: A Sample Sieve Approach

141、R-GAP: Recursive Gradient Attack on Privacy

142、Unsupervised Audiovisual Synthesis via Exemplar Autoencoders

143、Exploring Balanced Feature Spaces for Representation Learning

144、DOP: Off-Policy Multi-Agent Decomposed Policy Gradients

145、On the Bottleneck of Graph Neural Networks and its Practical Implications

146、NBDT: Neural-Backed Decision Tree

147、Shapley Explanation Networks

148、Diverse Video Generation using a Gaussian Process Trigger

149、Auto Seg-Loss: Searching Metric Surrogates for Semantic Segmentation

150、NeMo: Neural Mesh Models of Contrastive Features for Robust 3D Pose Estimation

151、MiCE: Mixture of Contrastive Experts for Unsupervised Image Clustering

152、Stochastic Security: Adversarial Defense Using Long-Run Dynamics of Energy-Based Models

153、Rank the Episodes: A Simple Approach for Exploration in Procedurally-Generated Environments

154、Denoising Diffusion Implicit Models

155、BERTology Meets Biology: Interpreting Attention in Protein Language Models

156、Trajectory Prediction using Equivariant Continuous Convolution

157、Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization

158、Robust and Generalizable Visual Representation Learning via Random Convolutions

159、VA-RED$^2$: Video Adaptive Redundancy Reduction

160、Bowtie Networks: Generative Modeling for Joint Few-Shot Recognition and Novel-View Synthesis

161、What Can You Learn From Your Muscles? Learning Visual Representation from Human Interactions

162、Graph Traversal with Tensor Functionals: A Meta-Algorithm for Scalable Learning

163、Multi-Time Attention Networks for Irregularly Sampled Time Series

164、Conservative Safety Critics for Exploration

165、Knowledge distillation via softmax regression representation learning

166、OPAL: Offline Primitive Discovery for Accelerating Offline Reinforcement Learning

167、Remembering for the Right Reasons: Explanations Reduce Catastrophic Forgetting

168、A Trainable Optimal Transport Embedding for Feature Aggregation and its Relationship to Attention

169、RODE: Learning Roles to Decompose Multi-Agent Tasks

170、Learning perturbation sets for robust machine learning

171、$i$-Mix: A Domain-Agnostic Strategy for Contrastive Representation Learning

172、Probing BERT in Hyperbolic Spaces

173、Practical Massively Parallel Monte-Carlo Tree Search Applied to Molecular Design

174、Multi-Prize Lottery Ticket Hypothesis: Finding Accurate Binary Neural Networks by Pruning A Randomly Weighted Network

175、Unbiased Teacher for Semi-Supervised Object Detection

176、High-Capacity Expert Binary Networks

177、Wasserstein Embedding for Graph Learning

178、Hopper: Multi-hop Transformer for Spatiotemporal Reasoning

179、Training with Quantization Noise for Extreme Model Compression

180、Multi-Class Uncertainty Calibration via Mutual Information Maximization-based Binning

181、Deep Encoder, Shallow Decoder: Reevaluating Non-autoregressive Machine Translation

182、Neural Pruning via Growing Regularization

183、Hierarchical Reinforcement Learning by Discovering Intrinsic Options

184、Is Label Smoothing Truly Incompatible with Knowledge Distillation: An Empirical Study

185、DINO: A Conditional Energy-Based GAN for Domain Translation

186、Optimal Conversion of Conventional Artificial Neural Networks to Spiking Neural Networks

187、Efficient Continual Learning with Modular Networks and Task-Driven Priors

188、Federated Semi-Supervised Learning with Inter-Client Consistency & Disjoint Learning

189、CPR: Classifier-Projection Regularization for Continual Learning

190、GAN2GAN: Generative Noise Learning for Blind Denoising with Single Noisy Images

191、Self-supervised Representation Learning with Relative Predictive Coding

192、Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets

193、Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data

194、On Learning Universal Representations Across Languages

195、FOCAL: Efficient Fully-Offline Meta-Reinforcement Learning via Distance Metric Learning and Behavior Regularization

196、GraphCodeBERT: Pre-training Code Representations with Data Flow

197、Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bits

198、MoPro: Webly Supervised Learning with Momentum Prototypes

199、AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights

200、FedBN: Federated Learning on Non-IID Features via Local Batch Normalization

201、A Unified Approach to Interpreting and Boosting Adversarial Transferability

202、Learning Subgoal Representations with Slow Dynamics

203、Learning N:M Fine-grained Structured Sparse Neural Networks From Scratch

204、Active Contrastive Learning of Audio-Visual Video Representations

205、PseudoSeg: Designing Pseudo Labels for Semantic Segmentation

206、Shape-Texture Debiased Neural Network Training

207、Revisiting Dynamic Convolution via Matrix Decomposition

208、Contemplating Real-World Object Classification

209、Bag of Tricks for Adversarial Training

210、MALI: A memory efficient and reverse accurate integrator for Neural ODEs

211、Explainable Deep One-Class Classification

212、Universal Weakly Supervised Segmentation by Pixel-to-Segment Contrastive Learning

213、Policy-Driven Attack: Learning to Query for Hard-label Black-box Adversarial Examples

214、Rethinking Soft Labels for Knowledge Distillation: A Bias–Variance Tradeoff Perspective

215、DARTS-: Robustly Stepping out of Performance Collapse Without Indicators

216、AdaFuse: Adaptive Temporal Fusion Network for Efficient Action Recognition

217、BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstruction

218、Generating Furry Cars: Disentangling Object Shape and Appearance across Multiple Domains

219、Distributed Momentum for Byzantine-resilient Stochastic Gradient Descent

ICLR2021 Accept Oral With Code

1、Complex Query Answering with Neural Link Predictors

2、Co-Mixup: Saliency Guided Joint Mixup with Supermodular Diversity

3、End-to-end Adversarial Text-to-Speech

4、Share or Not? Learning to Schedule Language-Specific Capacity for Multilingual Translation

5、When Do Curricula Work?

6、Parrot: Data-Driven Behavioral Priors for Reinforcement Learning

7、An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

8、Self-training For Few-shot Transfer Across Extreme Task Differences

9、DiffWave: A Versatile Diffusion Model for Audio Synthesis

10、Learning Cross-Domain Correspondence for Control with Dynamics Cycle-Consistency

11、A Distributional Approach to Controlled Text Generation

12、Gradient Projection Memory for Continual Learning

13、Rethinking Attention with Performers

14、Neural Synthesis of Binaural Speech From Mono Audio

15、Dataset Condensation with Gradient Matching

16、Free Lunch for Few-shot Learning: Distribution Calibration

ICLR2021 Accept Spotlight With Code

1、Self-supervised Visual Reinforcement Learning with Object-centric Representations

2、MARS: Markov Molecular Sampling for Multi-objective Drug Discovery

3、The Intrinsic Dimension of Images and Its Impact on Learning

4、Orthogonalizing Convolutional Layers with the Cayley Transform

5、Learning Mesh-Based Simulation with Graph Networks

6、CPT: Efficient Deep Neural Network Training via Cyclic Precision

7、PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics

8、Memory Optimization for Deep Networks

9、Watch-And-Help: A Challenge for Social Perception and Human-AI Collaboration

10、HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark

11、Dataset Inference: Ownership Resolution in Machine Learning

12、Learning from Protein Structure with Geometric Vector Perceptrons

13、Fast Geometric Projections for Local Robustness Certification

14、Sharpness-aware Minimization for Efficiently Improving Generalization

15、Self-Supervised Policy Adaptation during Deployment

16、Data-Efficient Reinforcement Learning with Self-Predictive Representations

17、Dynamic Tensor Rematerialization

18、Model-Based Visual Planning with Self-Supervised Functional Distances

19、Regularization Matters in Policy Optimization - An Empirical Study on Continuous Control

20、Autoregressive Entity Retrieval

21、Neural Approximate Sufficient Statistics for Implicit Models

22、Tent: Fully Test-Time Adaptation by Entropy Minimization

23、Are Neural Rankers still Outperformed by Gradient Boosted Decision Trees?

24、A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference

25、Large Scale Image Completion via Co-Modulated Generative Adversarial Networks

26、Undistillable: Making A Nasty Teacher That CANNOT teach students

27、A Good Image Generator Is What You Need for High-Resolution Video Synthesis

28、Noise against noise: stochastic label noise helps combat inherent label noise

29、Quantifying Differences in Reward Functions

30、Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images

31、UPDeT: Universal Multi-agent RL via Policy Decoupling with Transformers

32、Mutual Information State Intrinsic Control

33、Long-tailed Recognition by Routing Diverse Distribution-Aware Experts

34、Uncertainty Sets for Image Classifiers using Conformal Prediction

35、Stabilized Medical Image Attacks

36、Sequential Density Ratio Estimation for Simultaneous Optimization of Speed and Accuracy

37、Improving Adversarial Robustness via Channel-wise Activation Suppressing