statistical learning theory
A framework addressing the theoretical underpinnings of machine learning algorithms and models, focusing on concepts of generalization, model complexity, and performance bounds, which informs the design and evaluation of AI systems.
- Attention-based clustering
- Multimodal Bandits: Regret Lower Bounds and Optimal Algorithms
- On Union-Closedness of Language Generation
- Stability and Sharper Risk Bounds with Convergence Rate $\tilde{O}(1/n^2)$
- The Generative Leap: Tight Sample Complexity for Efficiently Learning Gaussian Multi-Index Models
- Tight Generalization Bounds for Large-Margin Halfspaces
- Which Algorithms Have Tight Generalization Bounds?