algorithm design
The process of developing and analyzing algorithms that will perform specific tasks efficiently. In AI, this includes creating novel algorithms that enhance performance, scalability, and generalization of models.
- $O(\sqrt{T})$ Static Regret and Instance Dependent Constraint Violation for Constrained Online Convex Optimization
- $\texttt{STRCMP}$: Integrating Graph Structural Priors with Language Models for Combinatorial Optimization
- Accelerated Evolving Set Processes for Local PageRank Computation
- AlgoTune: Can Language Models Speed Up General-Purpose Numerical Programs?
- Asymptotic theory of SGD with a general learning-rate
- Avoiding exp(R) scaling in RLHF through Preference-based Exploration
- Bridging Arbitrary and Tree Metrics via Differentiable Gromov Hyperbolicity
- Crucible: Quantifying the Potential of Control Algorithms through LLM Agents
- Data-Dependent Regret Bounds for Constrained MABs
- Exploring Landscapes for Better Minima along Valleys
- Generating and Checking DNN Verification Proofs
- Improved Regret and Contextual Linear Extension for Pandora's Box and Prophet Inequality
- Learning Across the Gap: Hybrid Multi-armed Bandits with Heterogeneous Offline and Online Data
- No-Regret Online Autobidding Algorithms in First-price Auctions
- Non-Clairvoyant Scheduling with Progress Bars
- Non-Stationary Lipschitz Bandits
- Online Bilateral Trade With Minimal Feedback: Don’t Waste Seller’s Time
- Online Two-Stage Submodular Maximization
- Optimistic Online-to-Batch Conversions for Accelerated Convergence and Universality
- Robust Contextual Pricing
- Sharp Analysis for KL-Regularized Contextual Bandits and RLHF
- Taming Adversarial Constraints in CMDPs
- The Parameterized Complexity of Computing the VC-Dimension
- Tractable Multinomial Logit Contextual Bandits with Non-Linear Utilities