Parallel bandit architecture based on laser chaos for reinforcement learning.
Takashi UrushibaraNicolas ChauvetSatoshi KochiSatoshi SunadaKazutaka KannoAtsushi UchidaRyoichi HorisakiMakoto NarusePublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- distributed processing
- multi processor
- parallel processing
- level parallelism
- master slave
- state space
- shared memory
- processor array
- neural network
- multi core processors
- parallel architecture
- function approximation
- genetic algorithm
- laser beam
- learning algorithm
- multi agent
- processing elements
- management system
- distributed memory
- optimal policy
- particle swarm optimization
- computer architecture
- reinforcement learning algorithms
- multi agent systems
- parallel implementation
- hardware implementation
- bandit problems
- markov decision processes
- machine learning