Counterexample-Guided Strategy Improvement for POMDPs Using Recurrent Neural Networks.
Steven CarrNils JansenRalf WimmerAlexandru Constantin SerbanBernd BeckerUfuk TopcuPublished in: CoRR (2019)
Keyphrases
- recurrent neural networks
- neural network
- feed forward
- artificial neural networks
- recurrent networks
- complex valued
- neural model
- reservoir computing
- cascade correlation
- echo state networks
- long short term memory
- markov decision processes
- feedforward neural networks
- significant improvement
- hebbian learning
- nonlinear dynamic systems
- artificial intelligence
- state space
- reinforcement learning
- model checking
- long term
- decision making