Alert button
Picture for Matthew Landen

Matthew Landen

Alert button

Planning with Abstract Learned Models While Learning Transferable Subtasks

Dec 16, 2019
John Winder, Stephanie Milani, Matthew Landen, Erebus Oh, Shane Parr, Shawn Squire, Marie desJardins, Cynthia Matuszek

Figure 1 for Planning with Abstract Learned Models While Learning Transferable Subtasks
Figure 2 for Planning with Abstract Learned Models While Learning Transferable Subtasks
Figure 3 for Planning with Abstract Learned Models While Learning Transferable Subtasks
Figure 4 for Planning with Abstract Learned Models While Learning Transferable Subtasks

We introduce an algorithm for model-based hierarchical reinforcement learning to acquire self-contained transition and reward models suitable for probabilistic planning at multiple levels of abstraction. We call this framework Planning with Abstract Learned Models (PALM). By representing subtasks symbolically using a new formal structure, the lifted abstract Markov decision process (L-AMDP), PALM learns models that are independent and modular. Through our experiments, we show how PALM integrates planning and execution, facilitating a rapid and efficient learning of abstract, hierarchical models. We also demonstrate the increased potential for learned models to be transferred to new and related tasks.

* Accepted at AAAI-20, 9 pages 
Viaarxiv icon