Goal-Directed Planning for Habituated Agents by Active Inference Using a Variational Recurrent Neural Network.
Takazumi MatsumotoJun TaniPublished in: Entropy (2020)
Keyphrases
- goal directed
- recurrent neural networks
- automated planning
- multi agent systems
- multi agent
- neural network
- complex valued
- rational agents
- multiple agents
- feed forward
- decision theoretic
- autonomous robots
- autonomous agents
- recurrent networks
- multiagent systems
- intelligent agents
- artificial neural networks
- echo state networks
- software agents
- dynamic environments
- reactive agents
- long short term memory
- image segmentation
- hidden layer
- reservoir computing
- bayesian networks
- action selection
- sum product
- nonlinear dynamic systems
- feedforward neural networks
- decision making
- neural model
- coalition formation
- single agent
- multi agent planning
- real time
- mobile robot
- mobile agents
- knowledge base
- artificial intelligence
- dirichlet process mixture models
- back propagation