The growing trend of fledgling reinforcement learning systems making their way into real-world applications has been accompanied by growing concerns for their safety and robustness. In recent years, a variety of approaches have been put forward to address the challenges of safety-aware reinforcement learning; however, these methods often either require a handcrafted model of the environment to be provided beforehand, or that the environment is relatively simple and low-dimensional. We present a novel approach to safety-aware deep reinforcement learning in high-dimensional environments called latent shielding. Latent shielding leverages internal representations of the environment learnt by model-based agents to "imagine" future trajectories and avoid those deemed unsafe. We experimentally demonstrate that this approach leads to improved adherence to formally-defined safety specifications.
Do Androids Dream of Electric Fences? Safety-Aware Reinforcement Learning with Latent Shielding
2022-02-28
In: Proceedings of the Workshop on Artificial Intelligence Safety 2022 (SafeAI 2022). ceur-ws.org: Virtual. (2022)
Paper
Electronic Resource
English
DDC: | 629 |
Engineering Index Backfile | 1964
|Battery-operated electric fences
Engineering Index Backfile | 1945
Pedestrian safety device for crosswalks equipped with safety fences
European Patent Office | 2022
|