Late Breaking Results: Reinforcement Learning for Scalable Logic Optimization with Graph Neural Networks.

Xavier Timoneda Lukas Cavigelli

Published in: DAC (2021)

Keyphrases

neural network
reinforcement learning
function approximation
graph structure
optimization problems
pattern recognition
random walk
optimization algorithm
optimization method
markov decision processes

genetic algorithm
optimization process
directed graph
connected components
structured data
multi valued
artificial neural networks
model free
graph theoretic

learning classifier systems
graph mining
graph theory
highly non linear
modal logic
feed forward
global optimization
optimal policy
learning algorithm

neural nets
bipartite graph
graph matching
neural network model
graph representation
sufficient conditions
learning capabilities
particle swarm optimization
function approximators