Scalable Spike-and-Slab
Comments Accepted to ICML 2022. Open-source software in Python and R available at https://github.com/niloyb/ScaleSpikeSlab
期刊&会议
ICML
International Conference on Machine Learning · 会议 · Machine Learning
COLA: Consistent Learning with Opponent-Learning Awareness
Comments Accepted @ ICML 2022
Transformer Quality in Linear Time
Comments Accepted to the 39th International Conference on Machine Learning (ICML'22)
Private Adaptive Optimization with Side Information
Comments ICML 2022
Exact Optimal Accelerated Complexity for Fixed-Point Iterations
Comments ICML 2022 Long Talk
Hierarchical Neural Network Approaches for Long Document Classification
Comments Accepted at International Conference on Machine Learning and Computing (ICMLC) 2022
Black-Box Tuning for Language-Model-as-a-Service
Comments Accepted by ICML 2022. Camera-ready version
Robin Hood and Matthew Effects: Differential Privacy Has Disparate Impact on Synthetic Data
Journal ref Proceedings of the 39th International Conference on Machine Learning (ICML 2022)
A Policy Efficient Reduction Approach to Convex Constrained Deep Reinforcement Learning
Comments Reinforcement Learning for Real Life (RL4RealLife) Workshop in the 38th International Conference on Machine Learning, 2021
Contrastive Mixture of Posteriors for Counterfactual Inference, Data Integration and Fairness
Comments Published as a conference paper (long presentation) at ICML 2022
Learning Stable Classifiers by Transferring Unstable Features
Comments ICML 2022
Attentional Meta-learners for Few-shot Polythetic Classification
Comments Accepted at ICML 2022
To Smooth or Not? When Label Smoothing Meets Noisy Labels
Comments ICML 2022 (Long Presentation)
MPC-based Imitation Learning for Safe and Human-like Autonomous Driving
Comments Accepted at the 1st Workshop on Safe Learning for Autonomous Driving (SL4AD), co-located with the 39th International Conference on Machine Learning (ICML 2022)
AnyMorph: Learning Transferable Polices By Inferring Agent Morphology
Comments published at ICML 2022
Phasic Self-Imitative Reduction for Sparse-Reward Goal-Conditioned Reinforcement Learning
Comments Accepted at ICML 2022
Three Applications of Conformal Prediction for Rating Breast Density in Mammography
Comments Accepted to Workshop on Distribution-Free Uncertainty Quantification at ICML 2022
Measuring Representational Robustness of Neural Networks Through Shared Invariances
Comments Accepted for oral presentation at ICML 2022
Towards Galaxy Foundation Models with Hybrid Contrastive Learning
Comments Accepted at the ICML 2022 Workshop on Machine Learning for Astrophysics. Data: www.github.com/mwalmsley/pytorch-galaxy-datasets. Please reach out to share your labelled data - all contributions will be credited in future work
RankSim: Ranking Similarity Regularization for Deep Imbalanced Regression
Comments Accepted to ICML 2022
Mitigating Neural Network Overconfidence with Logit Normalization
Comments Accepted by ICML 2022
Graph-Coupled Oscillator Networks
Comments ICML 2022
Generalizing to New Physical Systems via Context-Informed Dynamics Model
Comments Accepted at ICML 2022
Correlation Clustering via Strong Triadic Closure Labeling: Fast Approximation Algorithms and Practical Lower Bounds
Comments ICML 2022
Inductive Biases and Variable Creation in Self-Attention Mechanisms
Comments v2: camera-ready revisions for ICML 2022
A Topological characterisation of Weisfeiler-Leman equivalence classes
Comments To appear in ICML's Workshop Topology, Algebra and Geometry in Machine Learning
Prototype-Anchored Learning for Learning with Imperfect Annotations
Comments ICML 2022
Input-agnostic Certified Group Fairness via Gaussian Parameter Smoothing
Journal ref ICML 2022
Optimistic Linear Support and Successor Features as a Basis for Optimal Policy Transfer
Comments Proceedings of the 39th International Conference on Machine Learning (ICML'22)
Score-Guided Intermediate Layer Optimization: Fast Langevin Mixing for Inverse Problems
Comments Accepted to ICML 2022. 32 pages, 9 Figures
