Title

Disentangling Exploration from Exploitation

Author(s)

Alessandro Lizzeri Alessandro Lizzeri (Princeton University and NBER)

Eran Shmaya Eran Shmaya ( State University of New York at Stony Brook)

Leeat Yariv Leeat Yariv (Princeton University, CEPR, and NBER)

Abstract

Starting from Robbins (1952), the literature on experimentation via multi-armed bandits has wed exploration and exploitation. Nonetheless, in many applications, agents’ exploration and exploitation need not be intertwined: a policymaker may assess new policies different than the status quo; an investor may evaluate projects outside her portfolio. We characterize the optimal experimentation policy when exploration and exploitation are disentangled in the case of Poisson bandits, allowing for general news structures. The optimal policy features complete learning asymptotically, exhibits lots of persistence, but cannot be identified by an index à la Gittins. Disentanglement is particularly valuable for intermediate parameter values.

Creation Date

2024-04

Section URL ID

Paper Number

334

URL

https://gceps.princeton.edu/wp-content/uploads/2024/06/wp334_Yariv_ExplorationExploitation.pdf

File Function

Jel

C73, D81, D83, O35

Keyword(s)

Exploration and Exploitation, Poisson Bandits

Suppress

false

Series

3