Reinforcement learning is a proven technique for an agent to learn a task. However, when learning a task using reinforcement learning, the agent cannot distinguish the characteristics of the environment from those of the task. This makes it harder to transfer skills between tasks in the same environment. Furthermore, this does not reduce risk when training for a new task. In this paper, we introduce an approach to decouple the environment characteristics from the task-specific ones, allowing an agent to develop a sense of survival. We evaluate our approach in an environment where an agent must learn a sequence of collection tasks, and show that decoupled learning allows for a safer utilization of prior knowledge.
@article{arxiv.1708.02838,
title = {Decoupled Learning of Environment Characteristics for Safe Exploration},
author = {Pieter Van Molle and Tim Verbelen and Steven Bohez and Sam Leroux and Pieter Simoens and Bart Dhoedt},
journal= {arXiv preprint arXiv:1708.02838},
year = {2017}
}
Comments
4 pages, 4 figures, ICML 2017 workshop on Reliable Machine Learning in the Wild