stochastic optimization
An optimization approach that uses random sampling and approximation techniques to find optimal solutions, often employed in large-scale machine learning where exact computations are infeasible.
- Adaptive Algorithms with Sharp Convergence Rates for Stochastic Hierarchical Optimization
- Conditional Gradient Methods with Standard LMO for Stochastic Simple Bilevel Optimization
- Differentially Private Bilevel Optimization: Efficient Algorithms with Near-Optimal Rates
- Distances for Markov chains from sample streams
- Enforcing Hard Linear Constraints in Deep Learning Models with Decision Rules
- Generating Informative Samples for Risk-Averse Fine-Tuning of Downstream Tasks
- Learning from A Single Markovian Trajectory: Optimality and Variance Reduction
- MESS+: Dynamically Learned Inference-Time LLM Routing in Model Zoos with Service Level Guarantees
- MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization
- Mechanism Design via the Interim Relaxation
- On the Optimal Construction of Unbiased Gradient Estimators for Zeroth-Order Optimization
- Scalable inference of functional neural connectivity at submillisecond timescales
- Stab-SGD: Noise-Adaptivity in Smooth Optimization with Stability Ratios
- Teaching Transformers to Solve Combinatorial Problems through Efficient Trial & Error