Introduces a measure-transformation-based surrogate loss for solver-free training in predict-then-optimize problems, with Fisher consistency and excess risk bounds.
Integrated condi tional estimation-optimization
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
Derives generalization bounds for perturbed surrogate policies in combinatorial optimization by decomposing excess risk into perturbation bias controlled by fan-crossing probability, statistical estimation error, and optimization error.
Establishes equivalence conditions between nested and joint risk assessments in contextual optimization, shows policy independence from contextual risk measure under conditions, and proves SAA consistency in RKHS.
citing papers explorer
-
A Solver-Free Training Method for Predict-then-Optimize
Introduces a measure-transformation-based surrogate loss for solver-free training in predict-then-optimize problems, with Fisher consistency and excess risk bounds.
-
Generalization Bounds of Surrogate Policies for Combinatorial Optimization Problems
Derives generalization bounds for perturbed surrogate policies in combinatorial optimization by decomposing excess risk into perturbation bias controlled by fan-crossing probability, statistical estimation error, and optimization error.
-
Risk-averse Decision Making with Contextual Information: Model, Sample Average Approximation, and Kernelization
Establishes equivalence conditions between nested and joint risk assessments in contextual optimization, shows policy independence from contextual risk measure under conditions, and proves SAA consistency in RKHS.