AI Paper Atlas

Research map / Bandits and online learning / Multi-armed bandits

Leveraging Observations in Bandits: Between Risks and Benefits

Andrei Lupu, Audrey Durand, Doina Precup

AAAI 2019 · Bandits and online learning › Multi-armed bandits · approach: Non-stationary and structured-reward bandits · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Learn with Target-UCB from a UCB target instead of standalone UCBTarget-UCB · 2-actions bandit, mu*=0.9cumulative regretImproved
Use greedy follower of the UCB target instead of Target-UCBlearner following a UCB target · 2-actions bandit, mu*=0.9cumulative regretImproved
Increase alpha of the alpha-optimal target (ablation)Target-UCB · 2-actions bandit, mu*=0.9, delta_a=0.1cumulative regretImproved
Replace greedy follower with Target-UCB when target is poor (low alpha)learner following alpha-optimal target · 2-actions bandit, alpha=0.001cumulative regretImproved
Target-UCB clique of 11 agents vs single UCBTarget-UCB clique · 2-actions bandit, mu*=0.5, delta_a=0.1cumulative regretImproved
Target-UCB clique of 11 agents vs UCB clique sharing full informationTarget-UCB clique · 2-actions bandit, mu*=0.5cumulative regretImproved
Target-UCB graphs of 20 agents vs single UCBTarget-UCB graphs · random 10-actions banditscumulative regretImproved
Vary graph structure (chain, loop, clique, random, small-world) (ablation)Target-UCB graphs · random 10-actions banditscumulative regretNo clear effect
Target-UCB graphs vs UCB clique of 20 agentsTarget-UCB graphs · random 10-actions banditscumulative regretMixed
Target-UCB learning from a single human targetTarget-UCB · human 2-actions bandit, mu*=0.6cumulative regretImproved
Clique of four Target-UCB agents vs clique of four humansTarget-UCB clique · human 2-actions bandit, mu*=0.6cumulative regretImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy