2017/01/01 by El Mahdi El Mhamdi, Mhamdi, El Mahdi El, Rachid Guerraoui +5 · 3 citations
Computer Science · Mathematics · #Artificial Intelligence (cs.AI) #Data Stream Mining Techniques #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Multiagent Systems (cs.MA) #Reinforcement Learning in Robotics #cs.AI #cs.LG #cs.MA #stat.ML
paper · pdf · doi:10.48550/arxiv.1704.02882
openalex publication_date 2017/01/01 · arxiv created 2017/05/22 · arxiv updated 2017/05/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
In reinforcement learning, agents learn by performing actions and observing their outcomes. Sometimes, it is desirable for a human operator to interrupt an agent in order to prevent dangerous situations from happening. Yet, as part of their learning process, agents may link these interruptions, that impact their reward, to specific states and deliberately avoid them. The situation is particularly challenging in a multi-agent context because agents might not only learn from their own past interruptions, but also from those of other agents. Orseau and Armstrong defined safe interruptibility for one learner, but their work does not naturally extend to multi-agent systems. This paper introduces dynamic safe interruptibility, an alternative definition more suited to decentralized learning problems, and studies this notion in two learning frameworks: joint action learners and independent learners. We give realistic sufficient conditions on the learning algorithm to enable dynamic safe interruptibility in the case of joint action learners, yet show that these conditions are not sufficient for independent learners. We show however that if agents can detect interruptions, it is possible to prune the observations to ensure dynamic safe interruptibility even for independent learners.