Graph Networks as Learnable Physics Engines for Inference and Control.
Alvaro Sanchez-GonzalezNicolas HeessJost Tobias SpringenbergJosh MerelMartin A. RiedmillerRaia HadsellPeter W. BattagliaPublished in: ICML (2018)
Keyphrases
- control system
- bayesian networks
- directed graph
- social networks
- computer science
- small world
- artificial intelligence
- graph theory
- fully connected
- probabilistic networks
- graph theoretic
- graph structure
- probabilistic inference
- complex networks
- graph layout
- average degree
- clique tree
- overlapping communities
- graph grammars
- directed edges
- graph representation
- graph partitioning
- undirected graph
- graph model
- network analysis
- bipartite graph