Overview

Branch and bound solves hard optimization and verification problems by splitting them into subproblems, using lower and upper bounds to discard regions that cannot improve the best known solution. Work spans mixed-integer programming, neural-network verification, geometric estimation, routing, and robust regression, with stronger relaxations and problem-specific bounds helping reduce search. Another major direction learns branching and node-selection policies, while parallel CPU–GPU search and hybrid exact–heuristic methods target larger instances. The central tension is to make search faster and more scalable without weakening the guarantees that distinguish exact methods.

Latest papers

CardsList

33 papers5 papers in the last four weeks, up 25% on the four weeks before. 0.0% of all new papers.

  1. DC-Reg: Globally Optimal Point Cloud Registration via Tight Bounding with Difference of Convex Programming

    Mar 26, 2026Wei Lian, Fei Ma, Hang Pan +2Point Cloud RegistrationNonconvex Optimization

  2. Global Truncated Loss Minimization for Robust and Threshold-Resilient Geometric Estimation

    Mar 16, 2026Tianyu Huang, Liangzu Peng, Xinyue Zhang +5Robust Loss FunctionsBranch and Bound

  3. Advances in Global Solvers for 3D Vision

    Feb 16, 2026Zhenjun Zhao, Heng Yang, Bangyan Liao +7Convex RelaxationGlobal Optimization