Finite-State Controllers of POMDPs using Parameter Synthesis.
Sebastian JungesNils JansenRalf WimmerTim QuatmannLeonore WintererJoost-Pieter KatoenBernd BeckerPublished in: UAI (2018)
Keyphrases
- reinforcement learning
- input parameters
- program synthesis
- planning under uncertainty
- partially observable markov decision processes
- approximate solutions
- markov decision processes
- partially observable
- belief state
- parameter space
- distributed constraint optimization
- data sets
- markov chain
- image sequences
- decision trees
- genetic algorithm
- machine learning