Proceedings of the AAAI Conference on Artificial Intelligence · 2020 · 145 citations · 25 references
ComputingComputer ScienceAlgorithmsArtificial IntelligenceOptimizationCombinatorial OptimizationDiscrete OptimizationExploration-exploitation TradeoffHeuristic SearchReinforcement LearningCombinatorial Optimization HeuristicsContinual Reinforcement LearningExploration In Reinforcement LearningGraph OptimizationMaximum Cut ApproximationObjective Function Formulation
Many real-world problems can be reduced to combinatorial optimization on a graph, where the subset or ordering of vertices that maximize some objective function must be found. With such tasks often NP-hard and analytically intractable, reinforcement learning (RL) has shown promise as a framework with which efficient heuristic methods to tackle these problems can be learned. Previous works construct the solution subset incrementally, adding one element at a time, however, the irreversible nature of this approach prevents the agent from revising its earlier decisions, which may be necessary given the complexity of the optimization task. We instead propose that the agent should seek to continuously improve the solution by learning to explore at test time. Our approach of exploratory combinatorial optimization (ECO-DQN) is, in principle, applicable to any combinatorial problem that can be defined on a graph. Experimentally, we show our method to produce state-of-the-art RL performance on the Maximum Cut problem. Moreover, because ECO-DQN can start from any arbitrary configuration, it can be combined with other search methods to further improve performance, which we demonstrate using a simple random search.
25
Optimization by Simulated Annealing
Scott Kirkpatrick, C. D. Gelatt, M.P. Vecchi · Science · 1983 · 44.6K citations
Human-level control through deep reinforcement learning
Volodymyr Mnih, Koray Kavukcuoglu, David Silver et al. · Nature · 2015 · 30K citations
Statistical mechanics of complex networks
Réka Albert, Albert-Ĺaszló Barabási · Reviews of Modern Physics · 2002 · 20.6K citations · Full text
Mastering the game of Go without human knowledge
David Silver, Julian Schrittwieser, Karen Simonyan et al. · Nature · 2017 · 9.2K citations
Engineering, Educational Psychology, Cognitive Psychology +7