← Optimization & Theory
Machine Learning › Optimization & Theory ›

Theory

4950 directly classified papers

Papers per year

Papers

Provably Fast Convergence of Independent Natural Policy Gradient for Markov Potential Games NIPS 2023
Compositional Policy Learning in Stochastic Control Systems with Formal Guarantees NIPS 2023
TD Convergence: An Optimization Perspective NIPS 2023
On Dynamic Programming Decompositions of Static Risk Measures in Markov Decision Processes NIPS 2023
Reducing Blackwell and Average Optimality to Discounted MDPs via the Blackwell Discount Factor NIPS 2023
Survival Instinct in Offline Reinforcement Learning NIPS 2023
A Theoretical Analysis of Optimistic Proximal Policy Optimization in Linear Markov Decision Processes NIPS 2023
Replicability in Reinforcement Learning NIPS 2023
Spectral Entry-wise Matrix Estimation for Low-Rank Reinforcement Learning NIPS 2023
The Curious Price of Distributional Robustness in Reinforcement Learning with a Generative Model NIPS 2023
Trade-off Between Efficiency and Consistency for Removal-based Explanations NIPS 2023
Transformers are uninterpretable with myopic methods: a case study with bounded Dyck grammars NIPS 2023
Characterizing Out-of-Distribution Error via Optimal Transport NIPS 2023
Stability Guarantees for Feature Attributions with Multiplicative Smoothing NIPS 2023
Abide by the law and follow the flow: conservation laws for gradient flows NIPS 2023
On the spectral bias of two-layer linear networks NIPS 2023
When Does Optimizing a Proper Loss Yield Calibration? NIPS 2023
Evaluating Neuron Interpretation Methods of NLP Models NIPS 2023
Approximate Heavy Tails in Offline (Multi-Pass) Stochastic Gradient Descent NIPS 2023
Dynamics of Finite Width Kernel and Prediction Fluctuations in Mean Field Neural Networks NIPS 2023
Low Tensor Rank Learning of Neural Dynamics NIPS 2023
What is the Inductive Bias of Flatness Regularization? A Study of Deep Matrix Factorization Models NIPS 2023
Deep Neural Collapse Is Provably Optimal for the Deep Unconstrained Features Model NIPS 2023
Leveraging the two-timescale regime to demonstrate convergence of neural networks NIPS 2023
SugarCrepe: Fixing Hackable Benchmarks for Vision-Language Compositionality NIPS 2023