000214524 001__ 214524
000214524 005__ 20180317092134.0
000214524 0247_ $$2doi$$a10.1007/s10107-017-1216-6
000214524 022__ $$a0025-5610
000214524 037__ $$aARTICLE
000214524 245__ $$aData-Driven Inverse Optimization with Incomplete Information
000214524 260__ $$c2018
000214524 269__ $$a2018
000214524 336__ $$aJournal Articles
000214524 500__ $$aAvailable from Optimization Online
000214524 520__ $$aIn 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.
000214524 6531_ $$aInverse optimization
000214524 6531_ $$aData-driven optimization
000214524 6531_ $$aDistributionally robust optimization
000214524 700__ $$0248134$$aMohajerin Esfahani, Peyman$$g179150
000214524 700__ $$0248584$$aShafieezadeh Abadeh, Soroosh$$g243894
000214524 700__ $$0249201$$aHanasusanto, Grani Adiwena$$g258502
000214524 700__ $$0247589$$aKuhn, Daniel$$g239987
000214524 773__ $$j167$$k1$$q191-234$$tMathematical Programming
000214524 8564_ $$uhttp://www.optimization-online.org/DB_HTML/2015/12/5244.html$$zURL
000214524 909CO $$ooai:infoscience.tind.io:214524$$particle$$pCDM
000214524 909C0 $$0252496$$pRAO$$xU12788
000214524 917Z8 $$x239987
000214524 917Z8 $$x239987
000214524 917Z8 $$x239987
000214524 917Z8 $$x239987
000214524 937__ $$aEPFL-ARTICLE-214524
000214524 973__ $$aEPFL$$rREVIEWED$$sPUBLISHED
000214524 980__ $$aARTICLE