Blog

What we've shipped.

Everything's on arXiv. Code is on GitHub or it doesn't count.

Filter

12 of 12 publications
Search index →
Oral ICML 2026 · 2026

Optimistic Online Newton with Sublinear Regret on Strongly Curved Losses

A second-order online learner that matches the regret of the best Newton step in hindsight, while remaining one-pass with O(d²) memory.

Leila Aydın, Maya Singh, Renat Ostrovsky

arXiv Read →
Paper ICML 2026 · 2026

Learned Cardinality with Distribution-Free Confidence

Calibrated cardinality estimation under workload drift — confidence intervals that hold without distributional assumptions.

Jihoon Park, Aylin Demir, Sasha Volkov

arXiv Read →