ICML 2019 accepted papers on Learning theory and algorithms
146 of the 774 papers accepted at ICML 2019 (18.9%) are on Learning theory and algorithms, compared with 23.9% at ICML 2018. They fall into 9 topics.
See ICML 2019 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 · 30 papers
- Data Shapley: Equitable Valuation of Data for Machine Learning (147 citations)
- TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzing (134 citations)
- Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoff (60 citations)
- Fairwashing: the risk of rationalization (49 citations)
- Explaining Deep Neural Networks with a Polynomial Time Algorithm for Shapley Value Approximation (47 citations)
- Generalized Linear Rule Models (22 citations)
- Robust Learning from Untrusted Sources (16 citations)
- Statistical Foundations of Virtual Democracy (16 citations)
- Active Embedding Search via Noisy Paired Comparisons (13 citations)
- On discriminative learning of prediction uncertainty (11 citations)
- Sublinear Time Nearest Neighbor Search over Generalized Weighted Space (11 citations)
- A Tree-Based Method for Fast Repeated Sampling of Determinantal Point Processes (10 citations)
and 18 more in the interactive map.
Matrix and tensor methods · 27 papers
- Learning Fast Algorithms for Linear Transforms Using Butterfly Factorizations (42 citations)
- Measurements of Three-Level Hierarchical Structure in the Outliers in the Spectrum of Deepnet Hessians (33 citations)
- Geometry and Symmetry in Short-and-Sparse Deconvolution (29 citations)
- Large-Scale Sparse Kernel Canonical Correlation Analysis (23 citations)
- Loss Landscapes of Regularized Linear Autoencoders (20 citations)
- Faster Algorithms for Binary Matrix Factorization (18 citations)
- Efficient Nonconvex Regularized Tensor Completion with Structure-aware Proximal Iterations (17 citations)
- On the Universality of Invariant Networks (12 citations)
- Greedy Orthogonal Pivoting Algorithm for Non-Negative Matrix Factorization (11 citations)
- Homomorphic Sensing (10 citations)
- Screening rules for Lasso with non-convex Sparse Regularizers (10 citations)
- Noisy Dual Principal Component Pursuit (9 citations)
and 15 more in the interactive map.
Kernels and regression theory · 17 papers
- On the Spectral Bias of Neural Networks (151 citations)
- A Kernel Theory of Modern Data Augmentation (100 citations)
- Analogies Explained: Towards Understanding Word Embeddings (54 citations)
- A Kernel Perspective for Regularizing Deep Neural Networks (34 citations)
- MONK Outlier-Robust Mean Embedding Estimation by Median-of-Means (14 citations)
- Rehashing Kernel Evaluation in High Dimensions (11 citations)
- Scalable Learning in Reproducing Kernel Krein Spaces (9 citations)
- A Multitask Multiple Kernel Learning Algorithm for Survival Analysis with Application to Cancer Biology (8 citations)
- Learning and Data Selection in Big Datasets (6 citations)
- Nearest Neighbor and Kernel Survival Analysis: Nonasymptotic Error Bounds and Strong Consistency Rates (6 citations)
- Regularization in directable environments with application to Tetris (4 citations)
- Towards a Unified Analysis of Random Fourier Features (4 citations)
and 5 more in the interactive map.
Sample complexity · 16 papers
- Sample-Optimal Parametric Q-Learning Using Linearly Additive Features (101 citations)
- Near optimal finite time identification of arbitrary linear dynamical systems (74 citations)
- Sorting Out Lipschitz Function Approximation (65 citations)
- Sublinear quantum algorithms for training linear and kernel-based classifiers (41 citations)
- Approximation and non-parametric estimation of ResNet-type convolutional neural networks (28 citations)
- Concentration Inequalities for Conditional Value at Risk (18 citations)
- Distributed Learning with Sublinear Communication (17 citations)
- Maximum Likelihood Estimation for Learning Populations of Parameters (17 citations)
- On the statistical rate of nonlinear recovery in generative models with heavy-tailed data (16 citations)
- Uniform Convergence Rate of the Kernel Density Estimator Adaptive to Intrinsic Volume Dimension (13 citations)
- Fast Rates for a kNN Classifier Robust to Unknown Asymmetric Label Noise (9 citations)
- Refined Complexity of PCA with Outliers (6 citations)
and 4 more in the interactive map.
Submodular and game algorithms · 16 papers
- Optimal Auctions through Deep Learning (90 citations)
- Submodular Maximization beyond Non-negativity: Guarantees, Fast Algorithms, and Applications (62 citations)
- Submodular Streaming in All Its Glory: Tight Approximation, Minimum Memory and Low Adaptive Complexity (48 citations)
- Finding Mixed Nash Equilibria of Generative Adversarial Networks (16 citations)
- Non-monotone Submodular Maximization with Nearly Optimal Adaptivity and Query Complexity (16 citations)
- Beyond Adaptive Submodularity: Approximation Guarantees of Greedy Policy with Adaptive Submodularity Ratio (14 citations)
- Optimal Continuous DR-Submodular Maximization and Applications to Provable Mean Field Inference (13 citations)
- Robust Influence Maximization for Hyperparametric Models (9 citations)
- Sequential Facility Location: Approximate Submodularity and Greedy Algorithm (8 citations)
- Categorical Feature Compression via Submodular Optimization (6 citations)
- Optimal Kronecker-Sum Approximation of Real Time Recurrent Learning (5 citations)
- Submodular Cost Submodular Cover with an Approximate Oracle (5 citations)
and 4 more in the interactive map.
Clustering · 15 papers
- COMIC: Multi-view Clustering Without Parameter Selection (123 citations)
- Fair k-Center Clustering for Data Summarization (75 citations)
- Scalable Fair Clustering (55 citations)
- Guarantees for Spectral Clustering with Fairness Constraints (43 citations)
- DBSCAN++: Towards fast and scalable density clustering (42 citations)
- Proportionally Fair Clustering (42 citations)
- Power k-Means Clustering (38 citations)
- A Better k-means++ Algorithm via Local Search (21 citations)
- Improved Parallel Algorithms for Density-Based Network Clustering (18 citations)
- Neural Collaborative Subspace Clustering (13 citations)
- Spectral Clustering of Signed Graphs via Matrix Power Means (12 citations)
- Supervised Hierarchical Clustering with Exponential Linkage (8 citations)
and 3 more in the interactive map.
Fairness · 12 papers
- Flexibly Fair Representation Learning by Disentanglement (123 citations)
- Fair Regression: Quantitative Definitions and Reduction-Based Algorithms (64 citations)
- Fairness without Harm: Decoupled Classifiers with Preference Guarantees (51 citations)
- Learning Optimal Fair Policies (43 citations)
- Fairness risk measures (42 citations)
- Fairness-Aware Learning for Continuous Attributes and Treatments (39 citations)
- The Implicit Fairness Criterion of Unconstrained Learning (32 citations)
- On the Long-term Impact of Algorithmic Decision Policies: Effort Unfairness and Feature Segregation through Social Learning (24 citations)
- Repairing without Retraining: Avoiding Disparate Impact with Counterfactual Distributions (21 citations)
- Training Well-Generalizing Classifiers for Fairness Metrics and Other Data-Dependent Constraints (15 citations)
- Making Decisions that Reduce Discriminatory Impacts (12 citations)
- Stable and Fair Classification (11 citations)
Optimal transport · 9 papers
- On Efficient Optimal Transport: An Analysis of Greedy and Accelerated Mirror Descent Algorithms (57 citations)
- Obtaining Fairness using Optimal Transport Theory (51 citations)
- Subspace Robust Wasserstein Distances (50 citations)
- Wasserstein of Wasserstein Loss for Learning Generative Models (22 citations)
- Linear-Complexity Data-Parallel Earth Mover’s Distance Approximations (16 citations)
- Geometric Losses for Distributional Learning (8 citations)
- On Scalable and Efficient Computation of Large Scale Optimal Transport (7 citations)
- A Gradual, Semi-Discrete Approach to Generative Network Training via Explicit Wasserstein Minimization (4 citations)
- The Wasserstein Transform (3 citations)
Combinatorial and search optimization · 4 papers
- Cheap Orthogonal Constraints in Neural Networks: A Simple Parametrization of the Orthogonal and Unitary Group (86 citations)
- CapsAndRuns: An Improved Method for Approximately Optimal Algorithm Configuration (16 citations)
- Finding Options that Minimize Planning Time (7 citations)
- Static Automatic Batching In TensorFlow (6 citations)
