Contextual Optimization Under Model Misspecification: A Tractable and Generalizable Approach
Omar Bennouna, Jiawei Zhang, Saurabh Amin, Asuman E. Ozdaglar
Abstract
Contextual optimization problems are prevalent in decision-making applications where historical data and contextual features are used to learn predictive models that inform optimal actions. However, practical applications often suffer from model misspecification due to incomplete knowledge of the underlying data-generating process, leading to suboptimal decisions. Existing approaches primarily address the well-specified case, leaving a critical gap in handling misspecified models. In this paper, we propose a novel Integrated Learning and Optimization (ILO) framework that explicitly accounts for model misspecification by introducing a tractable surrogate loss function with strong theoretical guarantees on generalizability, tractability, and optimality. Our surrogate loss aligns with the true decision performance objective, ensuring robustness to misspecification without imposing restrictive assumptions. The proposed approach effectively mitigates the challenges of non-convexity and non-smoothness in the target loss function, leading to efficient optimization procedures. We provide rigorous theoretical analysis and experimental validation, demonstrating superior performance compared to state-of-the-art methods. Our work offers a principled solution to the practically relevant challenge of model misspecification in contextual optimization.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Decision-Focused Learning with Directional Gradients cites An Exact Symbolic Reduction of Linear Smart Predict+Optimize to Mixed Integer Linear Programming
- Adapting to misspecification in contextual bandits with offline regression oracles cites Adapting to Misspecification in Contextual Bandits
Builds on6
- Adapting to Misspecification in Contextual BanditsDylan J. Foster, Claudio Gentile, Mehryar Mohri, Julian ZimmertNeurIPS 2020 · 111 citations
- Adapting to misspecification in contextual bandits with offline regression oraclesSanath Kumar Krishnamurthy, Vitor Hadad, Susan AtheyICML 2021 · 27 citations
- Decision-Focused Learning with Directional GradientsMichael Huang, Vishal GuptaNeurIPS 2024 · 25 citations
- An Exact Symbolic Reduction of Linear Smart Predict+Optimize to Mixed Integer Linear ProgrammingJihwan Jeong, Parth Jaggi, Andrew Butler, Scott SannerICML 2022 · 14 citations
- Maximum Optimality Margin: A Unified Approach for Contextual Linear Programming and Inverse Linear ProgrammingChunlin Sun, Shang Liu, Xiaocheng LiICML 2023 · 13 citations
Related papers
- Landscape Surrogate: Learning Decision Losses for Mathematical Optimization Under Partial InformationArman Zharmagambetov, Brandon Amos, Aaron M. Ferber, Taoan Huang et al.NeurIPS 2023 · 29 citations
- The Bias-Variance Tradeoff in Data-Driven Optimization: A Local Misspecification PerspectiveHaixiang Lan, Luofeng Liao, Adam N. Elmachtoub, Christian Kroer et al.NeurIPS 2025 · 4 citations
- Predict-then-Calibrate: A New Perspective of Robust Contextual LPChunlin Sun, Linyu Liu, Xiaocheng LiNeurIPS 2023 · 26 citations
- A Discretization Framework for Robust Contextual Stochastic OptimizationRares Cristian, Georgia PerakisICLR 2024
- From Inverse Optimization to Feasibility to ERMSaurabh Mishra, Anant Raj, Sharan VaswaniICML 2024 · 4 citations
