For multi-objective best arm identification with independent Gaussian rewards, the paper derives the asymptotic complexity and proposes MO-BAI, a linear-programming algorithm achieving it up to a 1+eta factor.
In particular, we set αt = p 2 ln (4M Kt2/δ) /t in Algorithm 3 of Even-Dar et al
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Optimal Multi-Objective Best Arm Identification with Fixed Confidence
For multi-objective best arm identification with independent Gaussian rewards, the paper derives the asymptotic complexity and proposes MO-BAI, a linear-programming algorithm achieving it up to a 1+eta factor.