首页|Study Results from University of Southampton Provide New Insights into Artificia l Intelligence (Lifetime Policy Reuse and the Importance of Task Capacity)
Study Results from University of Southampton Provide New Insights into Artificia l Intelligence (Lifetime Policy Reuse and the Importance of Task Capacity)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Data detailed on Artificial Intelligen ce have been presented. According to news originating from Southampton, United K ingdom, by NewsRx correspondents, research stated, "A longstanding challenge in artificial intelligence is lifelong reinforcement learning, where learners are given many tasks in sequence and must transfer knowledge between tasks while avo iding catastrophic forgetting." Our news journalists obtained a quote from the research from the University of S outhampton, "Policy reuse and other multi-policy reinforcement learning techniqu es can learn multiple tasks but may generate many policies. This paper presents two novel contributions, namely 1) Lifetime Policy Reuse, a modelagnostic polic y reuse algorithm that avoids generating many policies by optimising a fixed num ber of near-optimal policies through a combination of policy optimisation and ad aptive policy selection; and 2) the task capacity, a measure for the maximal num ber of tasks that a policy can accurately solve." According to the news editors, the research concluded: "Comparing two state-ofth e-art base-learners, the results demonstrate the importance of Lifetime Policy R euse and task capacity based pre-selection on an 18-task partially observable Pa cman domain and a Cartpole domain of up to 125 tasks." This research has been peer-reviewed.
SouthamptonUnited KingdomEuropeArt ificial IntelligenceEmerging TechnologiesMachine LearningReinforcement Lea rningUniversity of Southampton
NETL
NSTL
