Exact Algorithm
Exact algorithms aim to find optimal solutions to computationally complex problems, a crucial task across diverse fields like machine learning and operations research. Current research focuses on improving the efficiency of these algorithms through techniques such as integrating machine learning models (e.g., neural networks, attention mechanisms) to accelerate computations and employing advanced mathematical programming methods (e.g., integer linear programming, semidefinite programming) to solve larger and more challenging instances. These advancements are significant because they enable the solution of previously intractable problems, leading to more accurate models and better decision-making in various applications, from vehicle routing to machine learning model training.
Papers
Optimization meets Machine Learning: An Exact Algorithm for Semi-Supervised Support Vector Machines
Veronica Piccialli, Jan Schwiddessen, Antonio M. Sudoso
Exact Algorithms and Lowerbounds for Multiagent Pathfinding: Power of Treelike Topology
Foivos Fioravantes, Dušan Knop, Jan Matyáš Křišťan, Nikolaos Melissinos, Michal Opler