ICML 2022 accepted papers on Learning theory and algorithms
192 of the 1,233 papers accepted at ICML 2022 (15.6%) are on Learning theory and algorithms, compared with 17.3% at ICML 2021. They fall into 9 topics.
See ICML 2022 in the interactive map
Submitting to ICML? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Prediction and decision losses · 49 papers
- A Rigorous Study of Integrated Gradients Method and Extensions to Internal Neuron Attributions (20 citations)
- Decision-Focused Learning: Through the Lens of Learning to Rank (16 citations)
- Framework for Evaluating Faithfulness of Local Explanations (13 citations)
- Locally Sparse Neural Networks for Tabular Biomedical Data (11 citations)
- Robust Counterfactual Explanations for Tree-Based Ensembles (8 citations)
- Why Should I Trust You, Bellman? The Bellman Error is a Poor Replacement for Value Error (7 citations)
- A Psychological Theory of Explainability (6 citations)
- Fisher SAM: Information Geometry and Sharpness Aware Minimisation (6 citations)
- Hierarchical Shrinkage: Improving the accuracy and interpretability of tree-based models. (5 citations)
- RieszNet and ForestRiesz: Automatic Debiased Machine Learning with Neural Nets and Random Forests (5 citations)
- Sample Efficient Learning of Predictors that Complement Humans (5 citations)
- Conformal Prediction Sets with Limited False Positives (4 citations)
and 37 more in the interactive map.
Kernels and regression theory · 29 papers
- Model Agnostic Sample Reweighting for Out-of-Distribution Learning (9 citations)
- A new similarity measure for covariate shift with applications to nonparametric regression (6 citations)
- Fluctuations, Bias, Variance & Ensemble of Learners: Exact Asymptotics for Convex Losses in High-Dimension (6 citations)
- More Than a Toy: Random Matrix Models Predict How Real-World Neural Representations Generalize (6 citations)
- Neural Tangent Kernel Beyond the Infinite-Width Limit: Effects of Depth and Initialization (4 citations)
- Robustness Implies Generalization via Data-Dependent Generalization Bounds (3 citations)
- Supervised Learning with General Risk Functionals (3 citations)
- Does the Data Induce Capacity Control in Deep Learning? (2 citations)
- Failure and success of the spectral bias prediction for Laplace Kernel Ridge Regression: the case of low-dimensional data (2 citations)
- Functional Output Regression with Infimal Convolution: Exploring the Huber and $ε$-insensitive Losses (2 citations)
- Gaussian Process Uniform Error Bounds with Unknown Hyperparameters for Safety-Critical Applications (2 citations)
- Improved Convergence Rates for Sparse Approximation Methods in Kernel-Based Learning (2 citations)
and 17 more in the interactive map.
Sample complexity · 27 papers
- Robust Kernel Density Estimation with Median-of-Means principle (8 citations)
- On Well-posedness and Minimax Optimal Rates of Nonparametric Q-function Estimation in Off-policy Evaluation (5 citations)
- Predicting Out-of-Distribution Error with the Projection Norm (4 citations)
- Synergy and Symmetry in Deep Learning: Interactions between the Data, Model, and Inference Algorithm (4 citations)
- Understanding Dataset Difficulty with $\mathcal{V}$-Usable Information (4 citations)
- Bounding the Width of Neural Networks via Coupled Initialization A Worst Case Analysis (3 citations)
- Fast Relative Entropy Coding with A* coding (3 citations)
- On the Finite-Time Complexity and Practical Computation of Approximate Stationarity Concepts of Lipschitz Functions (3 citations)
- Efficient PAC Learning from the Crowd with Pairwise Comparisons (2 citations)
- Measure Estimation in the Barycentric Coding Model (2 citations)
- Sequential and Parallel Constrained Max-value Entropy Search via Information Lower Bound (2 citations)
- A query-optimal algorithm for finding counterfactuals (1 citation)
and 15 more in the interactive map.
Matrix and tensor methods · 22 papers
- Near-Exact Recovery for Tomographic Inverse Problems via Deep Learning (29 citations)
- Estimation in Rotationally Invariant Generalized Linear Models via Approximate Message Passing (11 citations)
- Proximal Denoiser for Convergent Plug-and-Play Optimization with Nonconvex Regularization (11 citations)
- Quantum-Inspired Algorithms from Randomized Numerical Linear Algebra (4 citations)
- Fully-Connected Network on Noncompact Symmetric Space and Ridgelet Transform based on Helgason-Fourier Analysis (3 citations)
- Efficient Low Rank Convex Bounds for Pairwise Discrete Graphical Models (2 citations)
- Log-Euclidean Signatures for Intrinsic Distances Between Unaligned Datasets (2 citations)
- Scalable MCMC Sampling for Nonsymmetric Determinantal Point Processes (2 citations)
- Generalized Leverage Scores: Geometric Interpretation and Applications (1 citation)
- Inductive Matrix Completion: No Bad Local Minima and a Fast Algorithm (1 citation)
- More Efficient Sampling for Tensor Decomposition With Worst-Case Guarantees (1 citation)
- Approximate Frank-Wolfe Algorithms over Graph-structured Support Sets
and 10 more in the interactive map.
Clustering · 19 papers
- An iterative clustering algorithm for the Contextual Stochastic Block Model with optimality guarantees (3 citations)
- Deep Squared Euclidean Approximation to the Levenshtein Distance for DNA Storage (3 citations)
- A Tighter Analysis of Spectral Clustering, and Beyond (2 citations)
- Why the Rich Get Richer? On the Balancedness of Random Partition Models (2 citations)
- Convergence and Recovery Guarantees of the K-Subspaces Method for Subspace Clustering (1 citation)
- Gradient Based Clustering (1 citation)
- Individual Preference Stability for Clustering (1 citation)
- A Random Matrix Analysis of Data Stream Clustering: Coping With Limited Memory Resources
- Adapting k-means Algorithms for Outliers
- Bregman Power k-Means for Clustering Exponential Family Data
- Correlation Clustering via Strong Triadic Closure Labeling: Fast Approximation Algorithms and Practical Lower Bounds
- Deep Safe Incomplete Multi-view Clustering: Theorem and Algorithm
and 7 more in the interactive map.
Fairness · 14 papers
- Achieving Fairness at No Utility Cost via Data Reweighing with Influence (6 citations)
- Active Sampling for Min-Max Fairness (6 citations)
- Fair Representation Learning through Implicit Path Alignment (5 citations)
- Learning fair representation with a parametric integral probability metric (4 citations)
- Understanding Instance-Level Impact of Fairness Constraints (4 citations)
- Active fairness auditing (3 citations)
- Input-agnostic Certified Group Fairness via Gaussian Parameter Smoothing (3 citations)
- Fair Generalized Linear Models with a Convex Penalty (2 citations)
- Metric-Fair Classifier Derandomization
- Selective Regression under Fairness Criteria
- Simple and near-optimal algorithms for hidden stratification and multi-group learning
- Fairness Interventions as (Dis)Incentives for Strategic Manipulation
and 2 more in the interactive map.
Optimal transport · 11 papers
- Re-evaluating Word Mover’s Distance (11 citations)
- Linear-Time Gromov Wasserstein Distances using Low Rank Couplings and Costs (8 citations)
- Debiaser Beware: Pitfalls of Centering Regularized Transport Maps (6 citations)
- On Transportation of Mini-batches: A Hierarchical Approach (4 citations)
- Entropic Gromov-Wasserstein between Gaussian Distributions (3 citations)
- Exact Optimal Accelerated Complexity for Fixed-Point Iterations (3 citations)
- Discrete Probabilistic Inverse Optimal Transport (2 citations)
- Unsupervised Ground Metric Learning Using Wasserstein Singular Vectors (2 citations)
- Distribution Regression with Sliced Wasserstein Kernels (1 citation)
- Order Constraints in Optimal Transport (1 citation)
- Batch Greenkhorn Algorithm for Entropic-Regularized Multimarginal Optimal Transport: Linear Rate of Convergence and Iteration Complexity
Submodular and game algorithms · 11 papers
- Deletion Robust Submodular Maximization over Matroids (4 citations)
- Fictitious Play and Best-Response Dynamics in Identical Interest and Zero-Sum Stochastic Games (4 citations)
- Information Discrepancy in Strategic Learning (3 citations)
- Kernelized Multiplicative Weights for 0/1-Polyhedral Games: Bridging the Gap Between Learning in Extensive-Form and Normal-Form Games (3 citations)
- Simplex Neural Population Learning: Any-Mixture Bayes-Optimality in Symmetric Zero-sum Games (1 citation)
- For Learning in Symmetric Teams, Local Optima are Global Nash Equilibria
- Robustness in Multi-Objective Submodular Optimization: a Quantile Approach
- Exact Learning of Preference Structure: Single-peaked Preferences and Beyond
- RUMs from Head-to-Head Contests
- Selling Data To a Machine Learner: Pricing via Costly Signaling
- Streaming Algorithm for Monotone k-Submodular Maximization with Cardinality Constraints
Combinatorial and search optimization · 10 papers
- Learning to Cut by Looking Ahead: Cutting Plane Selection via Imitation Learning (18 citations)
- Quant-BnB: A Scalable Branch-and-Bound Method for Optimal Decision Trees with Continuous Features (5 citations)
- On the Convergence of Inexact Predictor-Corrector Methods for Linear Programming (3 citations)
- Constrained Discrete Black-Box Optimization using Mixed-Integer Programming (2 citations)
- Probabilistic Bilevel Coreset Selection
- Solving Stackelberg Prediction Game with Least Squares Loss via Spherically Constrained Least Squares Reformulation
- A Multi-objective / Multi-task Learning Framework Induced by Pareto Stationarity
- An Exact Symbolic Reduction of Linear Smart Predict+Optimize to Mixed Integer Linear Programming
- Minimum Cost Intervention Design for Causal Effect Identification
- Neural Fisher Discriminant Analysis: Optimal Neural Network Embeddings in Polynomial Time
