Trending

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and Machine Learning and more.Subscribe

Machine Learning

Reinforcement learning for extremal combinatorics and graph theory

Constructions in combinatorics via neural networks

In this paper, we show how by using a reinforcement learning algorithm, the deepcross-entropy method, one can find explicit constructions and counterexamplesto several open conjectures in extremal combinatorics and graph theory.

Amongst the conjectures we refute are a question of brualdi and cao about maximizing permanents of pattern avoiding matrices, and several problems related to the adjacency and distance eigenvalues of graphs.

Authors

Adam Zsolt Wagner

Related Topics

Reinforcement learning algorithm

Deep cross-entropy method

Read the Paper

Read Paper

Trending

Bookmarks

◐ Recommended Follows

Founder 42papers.com. Also bui...

Vojtěch Létal

More

◐ Latest Activity

More

Reinforcement learning for extremal combinatorics and graph theory - 42Papers

Read Related Papers

Best Arm Identification with Safety Constraints

Uniform Convergence Rates for Lipschitz Learning on Graphs

Vulcan: A Machine Learning Approach to Solving Steiner Tree Problem in Graphs

WalkingTime: Dynamic Graph Embedding Using Temporal-Topological Flows

Offline Reinforcement Learning: Fundamental Barriers for Value Function Approximation

On the Provable Generalization of Recurrent Neural Networks

Graph structure of quantum spin networks for quantum annealing

A Framework to Plant and Hide Extremely Sparse Network Topologies

On the Existence of Universal Lottery Tickets

Policy Gradient for Continuous Time and Space Reinforcement Learning