Interpreting Neural Policies with Disentangled Tree Representations.
Tsun-Hsuan WangWei XiaoTim SeydeRamin M. HasaniDaniela RusPublished in: CoRR (2022)
Keyphrases
- tree structure
- distributed representations
- network architecture
- neural network
- fitted q iteration
- optimal policy
- internal representations
- tree nodes
- learning rules
- control policies
- reward function
- nonlinear predictive control
- tree models
- tree construction
- neural model
- tree structures
- leaf nodes
- spanning tree
- graph structure
- higher level
- multi dimensional
- nearest neighbor