What we work on
We organize around problems, not techniques. Each area carries the same commitment to provable guarantees and practical deployment.
Bandits and Online Learning
Sequential decision-making under partial information — multi-armed bandits, contextual bandits, and adversarial online learning with provable regret guarantees.
Autobidding, Ranking and Recommender Systems
Auction-time bidding, ad pacing, and online ranking under budget and ROI constraints. Bringing online learning to the systems that decide what billions of users see.
DBMS Optimization
Online query optimization, learned cardinality estimation, and adaptive index/buffer management — pushing online learning into the heart of modern database systems.
Optimization
Convex and non-convex optimization theory, lower bounds, and parameter-free methods. The structural foundations under everything else we build.
Out of the lab
New preprints, conference papers, and code releases.
- Paper 2026/04/12
VK-LSVD: A Large-Scale Industrial Dataset for Short-Video Recommendation
- Paper 2026/03/17
Practical MCTS-based Query Optimization: A Reproducibility Study and new MCTS algorithm for complex queries
- Paper 2026/03/02
Uncertainty Quantification of Click and Conversion Estimates for the Autobidding
- Journal 2026/02/05
Survey of Modern Smooth Optimization Algorithms with Comparison Oracle
Let's work together
We do science for both academia and industry, and we read every application. Whether you're a prospective student, a collaborator, or a company with a hard problem, we'd love to hear from you.