Uncertainty quantification via a memristor Bayesian deep neural network for risk-sensitive reinforcement learning.
Yudeng LinQingtian ZhangBin GaoJianshi TangPeng YaoChongxuan LiShiyu HuangZhengwu LiuYing ZhouYuyi LiuWenqiang ZhangJun ZhuHe QianHuaqiang WuPublished in: Nat. Mac. Intell. (2023)
Keyphrases
- risk sensitive
- decision theory
- quantum mechanics
- utility function
- reinforcement learning
- expected utility
- neural network
- optimal control
- model free
- markov decision processes
- decision theoretic
- markov decision problems
- function approximation
- control policies
- function approximators
- risk averse
- decision makers
- reinforcement learning algorithms
- decision making
- optimality criterion
- machine learning
- optimal policy
- dynamic programming
- conditional probabilities
- temporal difference
- state space
- probability distribution
- learning capabilities
- policy iteration
- artificial neural networks
- radial basis function
- genetic algorithm
- efficient optimization
- artificial intelligence
- multi agent
- theoretical framework
- partially observable
- bayesian networks
- linear programming
- learning algorithm
- decision problems