214524
20181203024106.0
0025-5610
10.1007/s10107-017-1216-6
doi
ARTICLE
Data-Driven Inverse Optimization with Incomplete Information
2018
2018
Journal Articles
Available from Optimization Online
In data-driven inverse optimization an observer aims to learn the preferences of an agent who solves a parametric optimization problem depending on an exogenous signal. Thus, the observer seeks the agent's objective function that best explains a historical sequence of signals and corresponding optimal actions. We formalize this inverse optimization problem as a distributionally robust program minimizing the worst-case risk that the estimated decision (i.e., the decision implied by a particular candidate objective) differs from the agent's actual response to a random signal. We show that our framework offers attractive out-of-sample performance guarantees for different prediction errors and that the emerging inverse optimization problems can be reformulated as (or approximated by) tractable convex programs when the prediction error is measured in the space of objective values. A main strength of the proposed approach is that it naturally generalizes to situations where the observer has imperfect information, e.g., when the agent's true objective function is not contained in the space of candidate objectives, when the agent suffers from bounded rationality or implementation errors, or when the observed signal-response pairs are corrupted by measurement noise.
Inverse optimization
Data-driven optimization
Distributionally robust optimization
Mohajerin Esfahani, Peyman
179150
248134
Shafieezadeh Abadeh, Soroosh
243894
248584
Hanasusanto, Grani Adiwena
258502
249201
Kuhn, Daniel
239987
247589
191-234
1
Mathematical Programming
167
URL
http://www.optimization-online.org/DB_HTML/2015/12/5244.html
RAO
252496
U12788
oai:infoscience.tind.io:214524
article
CDM
239987
239987
239987
239987
EPFL-ARTICLE-214524
EPFL
PUBLISHED
REVIEWED
ARTICLE