empirical performance
Performance metrics obtained through practical experiments and real-world data rather than theoretical or simulated predictions.
- $\text{G}^2\text{M}$: A Generalized Gaussian Mirror Method to Boost Feature Selection Power
- A Few Moments Please: Scalable Graphon Learning via Moment Matching
- Accelerating data-driven algorithm selection for combinatorial partitioning problems
- Adaptive Data Analysis for Growing Data
- Asymptotic theory of SGD with a general learning-rate
- Automaton Constrained Q-Learning
- Bilevel ZOFO: Efficient LLM Fine-Tuning and Meta-Training
- Block-Biased Mamba for Long-Range Sequence Processing
- Constrained Posterior Sampling: Time Series Generation with Hard Constraints
- Coreset for Robust Geometric Median: Eliminating Size Dependency on Outliers
- Direct Fisher Score Estimation for Likelihood Maximization
- Equivariance Everywhere All At Once: A Recipe for Graph Foundation Models
- Fading to Grow: Growing Preference Ratios via Preference Fading Discrete Diffusion for Recommendation
- Gaussian Process Upper Confidence Bound Achieves Nearly-Optimal Regret in Noise-Free Gaussian Process Bandits
- HyperMARL: Adaptive Hypernetworks for Multi-Agent RL
- Improved Balanced Classification with Theoretically Grounded Loss Functions
- Instance-Optimality for Private KL Distribution Estimation
- Integral Imprecise Probability Metrics
- Intrinsic Benefits of Categorical Distributional Loss: Uncertainty-aware Regularized Exploration in Reinforcement Learning
- Learning-Augmented Algorithms for $k$-median via Online Learning
- Low-Rank Graphon Learning for Networks
- MeCeFO: Enhancing LLM Training Robustness via Fault-Tolerant Optimization
- Mean Flows for One-step Generative Modeling
- Mean Flows for One-step Generative Modeling
- Model Selection for Off-policy Evaluation: New Algorithms and Experimental Protocol
- Model-Informed Flows for Bayesian Inference
- Multi-Class Support Vector Machine with Differential Privacy
- New Perspectives on the Polyak Stepsize: Surrogate Functions and Negative Results
- On the Hardness of Conditional Independence Testing In Practice
- Optimal Dynamic Regret by Transformers for Non-Stationary Reinforcement Learning
- Physics-Constrained Flow Matching: Sampling Generative Models with Hard Constraints
- Preserving Task-Relevant Information Under Linear Concept Removal
- Private Continual Counting of Unbounded Streams
- Rethinking Gradient Step Denoiser: Towards Truly Pseudo-Contractive Operator
- SPACE: Noise Contrastive Estimation Stabilizes Self-Play Fine-Tuning for Large Language Models
- STAR-Bets: Sequential TArget-Recalculating Bets for Tighter Confidence Intervals
- Sketched Gaussian Mechanism for Private Federated Learning
- Statistical Analysis of an Adversarial Bayesian Weak Supervision Method
- Stochastic Optimization in Semi-Discrete Optimal Transport: Convergence Analysis and Minimax Rate
- Stochastically Dominant Peer Prediction
- Streaming Stochastic Submodular Maximization with On-Demand User Requests
- The Gaussian Mixing Mechanism: Renyi Differential Privacy via Gaussian Sketches
- Through the River: Understanding the Benefit of Schedule-Free Methods for Language Model Training
- Understanding Adam Requires Better Rotation Dependent Assumptions
- Value Improved Actor Critic Algorithms
- What Expressivity Theory Misses: Message Passing Complexity for GNNs