{"ID":2846868,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.02048","arxiv_id":"2511.02048","title":"Finding Probably Approximate Optimal Solutions by Training to Estimate the Optimal Values of Subproblems","abstract":"The paper is about developing a solver for maximizing a real-valued function of binary variables. The solver relies on an algorithm that estimates the optimal objective-function value of instances from the underlying distribution of objectives and their respective sub-instances. The training of the estimator is based on an inequality that facilitates the use of the expected total deviation from optimality conditions as a loss function rather than the objective-function itself. Thus, it does not calculate values of policies, nor does it rely on solved instances.","short_abstract":"The paper is about developing a solver for maximizing a real-valued function of binary variables. The solver relies on an algorithm that estimates the optimal objective-function value of instances from the underlying distribution of objectives and their respective sub-instances. The training of the estimator is based o...","url_abs":"https://arxiv.org/abs/2511.02048","url_pdf":"https://arxiv.org/pdf/2511.02048v1","authors":"[\"Nimrod Megiddo\",\"Segev Wasserkrug\",\"Orit Davidovich\",\"Shimrit Shtern\"]","published":"2025-11-03T20:29:30Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false}