Graph networks as learnable physics engines for inference and control.
Alvaro Sanchez-GonzalezNicolas HeessJost Tobias SpringenbergJosh MerelMartin A. RiedmillerRaia HadsellPeter W. BattagliaPublished in: CoRR (2018)
Keyphrases
- probabilistic networks
- computer science
- control system
- probabilistic inference
- graph structures
- small world
- community discovery
- graph matching
- graph structure
- bayesian inference
- fully connected
- artificial intelligence
- graph representation
- overlapping communities
- bayesian networks
- graph layout
- dynamic networks
- graph databases
- efficient learning
- undirected graph
- control strategy
- directed graph
- structured data
- social networks