DualSMC: Tunneling Differentiable Filtering and Planning under Continuous POMDPs.
Yunbo WangBo LiuJiajun WuYuke ZhuSimon S. DuLi Fei-FeiJoshua B. TenenbaumPublished in: IJCAI (2020)
Keyphrases
- partially observable markov decision processes
- predictive state representations
- belief space
- partially observable
- reinforcement learning
- stochastic domains
- belief state
- planning problems
- sequential decision making problems
- continuous action
- continuous state spaces
- dynamical systems
- point based value iteration
- information filtering
- adaptive filtering
- objective function
- state space
- partial observability
- blocks world
- decision problems
- continuous state
- planning under uncertainty
- decision theoretic planning
- continuous domains
- heuristic search
- distributed constraint optimization
- filtering algorithm
- optimal planning
- partially observable markov decision process
- markov decision problems
- finite state
- np hard
- dynamic programming
- motion planning
- situation calculus
- dynamic environments
- denoising
- action selection
- decision theoretic
- planning domains
- action space
- planning systems
- image filtering
- ai planning
- markov decision processes