Quartz 4

Home

❯

ML

❯

Cumulative regret

Cumulative regret

Sep 23, 20251 min read

Metryka używana do oceny algorytmów z dziedziny Contextual Multi-arm Bandit Problem.

To jest po prostu suma Regret wszystkich wykonanych akcji.

Źródło: A Practical Method for Solving Contextual Bandit Problems Using Decision Trees


Graph View

Backlinks

  • Contextual Multi-arm Bandit Problem
  • Linear Thompson sampling, Contextual Bandits
  • Thompson Sampling for Contextual Bandits with Linear Payoffs

Created with Quartz v4.4.1 © 2025

  • GitHub
  • Discord Community