Skip to main content
Search

# Multi-fidelity tree bandit

1 related articles

Subscribe to updates
  1. arXiv

    CANOPY uses random-path probes to certify smoothness violations online, improving routing, top-k identification, test-time search, caching, and prompt trimming at matched budgets

    The work introduces CANOPY, a multi-fidelity tree bandit that uses cheap random-path probes to build an online certificate of local aggregation bias and directs expensive leaf evaluations toward cells where the certificate detects a smoothness violation, rather than assuming a global smoothness prior; the authors prove fixed-budget and regret guarantees whose additional cost is additive in the number of discontinuities, recovering the smooth-tree rate when no violations are present and approaching structure-blind search as violations become dense; across routing, top-k identification, test-time search, caching, and prompt trimming, CANOPY consistently improves matched-budget performance, including 2.9x higher top-10 recall on a 1000-model pool, 1.

You’re all caught up