
CaVE proposes a novel approach for end-to-end training in binary linear programs, aligning cost vectors to optimal subcones efficiently.


coremsg

cave-a-cone-aligned-approach-for-fast-predict-then-optimize-with-binary-linear-programs


CaVE: A Cone-Aligned Approach for Fast Predict-then-optimize with Binary Linear Programs


title_rewrite


CaVE proposes a novel approach for efficient end-to-end training of ML models to predict cost coefficients of binary linear optimization problems, achieving a favorable trade-off between training time and solution quality.
