Data-driven inverse optimization with imperfect information
From MaRDI portal
Publication:681497
DOI10.1007/S10107-017-1216-6zbMATH Open1406.90087arXiv1512.05489OpenAlexW2196462312MaRDI QIDQ681497FDOQ681497
Authors: Peyman Mohajerin Esfahani, Soroosh Shafieezadeh-Abadeh, Grani A. Hanasusanto, Daniel Kuhn 
Publication date: 12 February 2018
Published in: Mathematical Programming. Series A. Series B (Search for Journal in Brave)
Abstract: 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 focus here on situations where the observer has imperfect information, that is, where the agent's true objective function is not contained in the search space of candidate objectives, where the agent suffers from bounded rationality or implementation errors, or where the observed signal-response pairs are corrupted by measurement noise. We formalize this inverse optimization problem as a distributionally robust program minimizing the worst-case risk that the {em predicted} decision ({em i.e.}, the decision implied by a particular candidate objective) differs from the agent's {em actual} response to a random signal. We show that our framework offers rigorous out-of-sample guarantees for different loss functions used to measure prediction errors and that the emerging inverse optimization problems can be exactly reformulated as (or safely approximated by) tractable convex programs when a new suboptimality loss function is used. We show through extensive numerical tests that the proposed distributionally robust approach to inverse optimization attains often better out-of-sample performance than the state-of-the-art approaches.
Full work available at URL: https://arxiv.org/abs/1512.05489
Recommendations
- Inverse optimization with noisy data
- Inverse integer optimization with multiple observations
- Inverse optimization for the recovery of constraint parameters
- Inverse multiobjective optimization: inferring decision criteria from data
- Imputing a variational inequality function or a convex objective function: a robust approach
Convex programming (90C25) Stochastic programming (90C15) Minimax problems in mathematical programming (90C47)
Cites Work
- The elements of statistical learning. Data mining, inference, and prediction
- Title not available (Why is that?)
- Some NP-complete problems in quadratic and nonlinear programming
- Robust optimization
- Inverse combinatorial optimization: a survey on problems, methods, and results
- On the rate of convergence in Wasserstein distance of the empirical measure
- On general minimax theorems
- Finite-dimensional variational inequality and nonlinear complementarity problems: A survey of theory, algorithms and applications
- Estimating Dynamic Models of Imperfect Competition
- Inverse Optimization
- Efficient Algorithms for the Inverse Spanning-Tree Problem
- Simple bounds for the convergence of empirical and occupation measures in 1-Wasserstein distance
- On an instance of the inverse shortest paths problem
- Computational Difficulties of Bilevel Linear Programming
- Convex optimization: algorithms and complexity
- A numerical procedure for inferring from experimental data the optimization cost functions using a multibody model of the neuro-musculoskeletal system
- Inverse optimization for linearly constrained convex separable programming problems
- Inverse conic programming with applications
- A Faster Algorithm for the Inverse Spanning Tree Problem
- The inverse optimal value problem
- Inverse integer programming
- The inverse newsvendor problem: choosing an optimal demand portfolio for capacitated resources
- An analytical approach to the problem of inverse optimization with additive objective functions: An application to human prehension
- Cutting plane algorithms for the inverse mixed integer linear programming problem
- Inverse optimization in high-speed networks
- Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations
- Generalized inverse multiobjective optimization with application to cancer therapy
- Behavioral Estimation of Mathematical Programming Objective Function Coefficients
- Inverse optimization: a new perspective on the Black-Litterman model
- Inverse optimization with noisy data
- Regularization via mass transportation
- Data-driven estimation in equilibrium using inverse optimization
Cited In (29)
- On correcting inputs: inverse optimization for online structured prediction
- Inverse Mixed Integer Optimization: Polyhedral Insights and Trust Region Methods
- Title not available (Why is that?)
- Inverse optimization with noisy data
- Statistics with set-valued functions: applications to inverse approximate optimization
- Data-driven optimization model customization
- Maximum likelihood estimation by Monte Carlo simulation: toward data-driven stochastic modeling
- Inverse optimization for the recovery of constraint parameters
- Inferring linear feasible regions using inverse optimization
- Robust inverse optimization
- Using inverse optimization to learn cost functions in generalized Nash games
- Inverse multiobjective optimization: inferring decision criteria from data
- Objective selection for cancer treatment: an inverse optimization approach
- A survey of network interdiction models and algorithms
- Optimality-based clustering: an inverse optimization approach
- Distributionally Favorable Optimization: A Framework for Data-Driven Decision-Making with Endogenous Outliers
- Imputing a variational inequality function or a convex objective function: a robust approach
- Quantile inverse optimization: improving stability in inverse linear programming
- Trajectory modeling via random utility inverse reinforcement learning
- Inverse attribute‐based optimization with an application in assortment optimization
- Inverse optimization with kernel regression: application to the power forecasting and bidding of a fleet of electric vehicles
- Inverse integer optimization with an imperfect observation
- Inverse integer optimization with multiple observations
- Data-driven decisions for problems with an unspecified objective function
- Decomposition and Adaptive Sampling for Data-Driven Inverse Linear Optimization
- Data-driven approximate value iteration with optimality error bound analysis
- Inverse Bayesian optimization: learning human acquisition functions in an exploration vs exploitation search task
- Frameworks and results in distributionally robust optimization
- Data-driven estimation in equilibrium using inverse optimization
Uses Software
This page was built for publication: Data-driven inverse optimization with imperfect information
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q681497)
