NeurIPS 2026 accepted papers on Learning theory and algorithms
376 of the 6,080 papers accepted at NeurIPS 2026 (6.2%) are on Learning theory and algorithms, compared with 7.0% at NeurIPS 2025. They fall into 9 topics.
See NeurIPS 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Prediction and decision losses · 98 papers
- A Kernel Nonconformity Score for Multivariate Conformal Prediction
- A Statistical Framework for Algorithmic Collective Action with Multiple Collectives
- Amortized Linear-time Exact Shapley Value for Product-Kernel Methods
- Class Adaptive Conformal Training
- Cluster with Auctions for Vector Search
- Co-optimization for Adaptive Conformal Prediction
- Conformal Agent Error Attribution
- Conformal Selective Acting: Anytime-Valid Risk Control for RLVR-Trained LLMs
- Counterfactual Maps: What They Are and How to Find Them
- Counterfactual Online Conformal Prediction Under Adaptive Logging
- Crossing the Validation crisis: Cross-validation reduces benchmarking variance surprisingly well
- Flow-Based Conformal Predictive Distributions
and 86 more in the interactive map.
Sample complexity · 59 papers
- Asymptotic Anytime-Valid Inference for U-statistics
- Autoregressive Learning in Joint KL: Sharp Oracle Bounds and Lower Bounds
- Bayes-Optimal BER and AUC: Estimation and Evaluation of Estimators
- Contrastive Identification and Generation in the Limit
- Efficient Learning of Truncated Boolean Product Distributions: Influence to the Rescue
- Group Distributionally Robust Optimization with Flexible Sample Queries
- Instance-Optimal Estimation with Multiple LLM Judges on a Budget
- Iterative Chow Filtering for Learning with Distribution Shift
- Learning Distributions from Multiple Data Providers
- MaxSketch: Robust Distinct Counting in Streams via Random Projections
- Mean Testing under Truncation beyond Gaussian
- Model Capacity Determines Grokking through Competing Memorisation and Generalisation Speeds
and 47 more in the interactive map.
Kernels and regression theory · 47 papers
- A mathematical theory of balancing relational generalization and memorization
- A Measure-Theoretic Analysis of Reasoning: Structural Generalization and Approximation Limits
- Certification-Enhanced Generalization Bounds
- Characterizing the Generalization Error of Random Feature Regression with Arbitrary Data-Augmentation
- Distribution Shift in Missing Data Imputation: A Risk-Based Perspective and Importance-Weighted Correction under MAR
- Gaussian Mixture Models in Hilbert Spaces via Kernel Methods
- Generalization at the Edge of Stability
- Intrinsic-dimension empirical Bernstein inequalities for bounded self-adjoint operators
- Kernelized Activation Steering
- Learning with Synthetic Data via SGD in High-Dimensional Linear Regression
- Length Generalization for Transformers via Compression
- Manifold Random Features
and 35 more in the interactive map.
Matrix and tensor methods · 38 papers
- A Novel Schur-Decomposition-Based Weight Projection Method for Stable State-Space Neural-Network Architectures
- Building Transformation Layers for Riemannian Neural Networks
- Debiasing Random Oblique Projections for Subsampled OLS and Fast CUR in High Dimensions
- Fast and Stable Gradient Approximation for Bilinear Forms of Hermitian Matrix Functions
- Finding Koopman Invariant Subspaces via Personalized PageRank
- Free Decompression with Algebraic Spectral Curves
- From Density Matrices to Phase Transitions in Deep Learning: Spectral Early Warnings and Interpretability
- Group-Aware Matrix Estimation and Latent Subspace Recovery
- High-Probability Minimax Adaptive Estimation in Besov Spaces via Online-to-Batch
- Intrinsic Muon: Spectral Optimization on Riemannian Matrix Manifolds
- Intrinsic Riemannian Cross-covariance for Manifold-valued Random Objects
- Learning to Discover Iterative Spectral Algorithms
and 26 more in the interactive map.
Combinatorial and search optimization · 34 papers
- Goal-Conditioned Supervised Learning for Multi-Objective Recommendation (1 citation)
- Position: Adopt Constraints Over Fixed Penalties in Deep Learning (1 citation)
- An Efficient Algorithm for Thresholding Monte Carlo Tree Search
- B$^3$-PWL: GPU-Batched Branch-and-Bound for Piecewise-Linear Optimization with SOS2 Constraints
- Differentiable Learning of Lifted Action Schemas for Classical Planning
- Differentiable Nonlinear Model Predictive Control
- Graph-Based Stochastic-Power-UCT: Monte-Carlo Graph Search with Power Mean Estimation
- MAdam: Metric-Aware Multi-Objective Adam
- Mixture-Trained Merging for Unified Multi-Objective Models
- On the Meta-Design of Allocation Problems
- Regularized Large Neighborhood Search
- SOC-ICNN: From Polyhedral to Conic Geometry for Learning Convex Surrogate Functions
and 22 more in the interactive map.
Submodular and game algorithms · 30 papers
- Approximate Envy-Free Allocations up to any k Goods
- Beyond the Half Approximation: Fair and Efficient Online Class Matching
- Epistemic Pairwise Maximin Share
- Incremental Multiple Oracle
- Non-Linear Pricing Restores Tractability for a Data Seller
- Online Fair Division Meets Reordering Buffers
- Optimizing the Envy Cycle Elimination Algorithm
- Paradoxes of Game Theoretic Equilibria and Price of Anarchy
- Polynomial-Time Algorithm for Thiele Voting Rules with Voter Interval Preferences
- Robust PAC Learning of Concurrent Stochastic Games
- Watermarking Game-Playing Agents in Perfect-Information Extensive-Form Games
- Compute Allocation Under Model-Provider Competition
and 18 more in the interactive map.
Optimal transport · 28 papers
- A Biconvex Formulation for Stable Transport of Mixture Models with a Unique Solution
- A Provably Convergent and Practical Algorithm for Gromov–Wasserstein Optimal Transport
- A Unified Approach for Computing Wasserstein Barycenters of Discrete and Continuous Measures
- Amortized Optimal Transport from Sliced Potentials
- Beyond Empirical Support: Structured Outlier Generation via Sinkhorn Optimal Transport
- Beyond Kemeny Medians: Consensus Ranking Distributions. Definition, Properties and Statistical Learning
- Block-Wise Differentiable Sinkhorn Attention: Tail-Refinement Gradients with a Gap-Aware Dustbin Bridge
- Conditional Optimal Bridge for Riemannian Activation Steering
- Efficient Transferable Optimal Transport via Min-Sliced Transport Plans
- Fréchet Regression on the Bures-Wasserstein Manifold
- Geometry-Aware Optimal Transport: Fast Intrinsic Dimension and Wasserstein Distance Estimation
- Log-Averaged Mirror Prox for Fast, Large-Scale Optimal Transport in Linear Space
and 16 more in the interactive map.
Fairness · 21 papers
- A Comprehensive View of Fairness through Distributional Stability
- Accuracy vs. Accuracy: Computational Tradeoffs Between Classification Rates and Utility
- Cost Efficient Fairness Audit Under Partial Feedback
- Demographic parity in regression and classification within the unawareness framework
- FairMT: Fairness for Heterogeneous Multi-Task Learning
- Geometry of Relaxed Fair Regression: A Unified Framework for Aware and Unaware Settings
- Intersectional Fairness via Mixed-Integer Optimization
- Scalable Fair Learning via Cramér-von Mises Regularization
- The Panel Complexity of Sortition: Is 12 Angry Men Enough?
- An Efficient Geometric Characterization of Robust Fair Learning
- An Information-theoretic Framework for Auditing Unfairness in Training Data
- Elicited Adaptation: Auditable Localized Fairness via Pairwise Queries
and 9 more in the interactive map.
Clustering · 21 papers
- Beyond Worst-Case Coreset Bounds for $k$-Clustering via Determinantal Sampling
- Classification Fields: Arbitrarily Fine Recursive Hierarchical Clustering From Few Examples
- Dynamic k-center clustering with lifetimes
- Robust Multi-view Clustering against Imperfect Information
- ToMAToMP: Robust and Multi-Parameter Topological Clustering
- A Batched Hartigan's $k$-Means Clustering
- Atoms and Pivot Based Correlation Clustering
- Coresets for Clustering Using Noisy Comparisons and Few Distance Queries
- Fair Range k-Supplier Clustering in Offline and Streaming Models
- Faster Dynamic Graph Clustering with Hierarchical Graph Contraction
- Fed-AGA: An Anchor Graph Alignment Framework for Federated Unaligned Multi-view Clustering
- Hierarchical Agglomerative Clustering via Relaxed Representatives
and 9 more in the interactive map.