Learnings Options End-to-End for Continuous Action Tasks
arXiv:1712.00004
Abstract
We present new results on learning temporally extended actions for continuoustasks, using the options framework (Suttonet al.[1999b], Precup [2000]). In orderto achieve this goal we work with the option-critic architecture (Baconet al.[2017])using a deliberation cost and train it with proximal policy optimization (Schulmanet al.[2017]) instead of vanilla policy gradient. Results on Mujoco domains arepromising, but lead to interesting questions aboutwhena given option should beused, an issue directly connected to the use of initiation sets.
Cited by in corpus (5)
- Interpretable Reinforcement Learning Inspired by Piaget's Theory of Cognitive Development
- Learning to Locomote with Deep Neural-Network and CPG-based Control in a Soft Snake Robot
- oIRL: Robust Adversarial Inverse Reinforcement Learning with Temporally Extended Actions
- DynoPlan: Combining Motion Planning and Deep Neural Network based Controllers for Safe HRL
- Options of Interest: Temporal Abstraction with Interest Functions