We present several completely general diversity metrics to quantify the problem-solving capacity of any public policy decision making process. This is performed by modelling the policy process using a declarative process paradigm in conjunction with constraints modelled by expressions in linear temporal logic. We introduce a class of traces, called first-passage traces, to represent the different executions of the declarative processes. Heuristics of what properties a diversity measure of such processes ought to satisfy are used to derive two different metrics for these processes in terms of the set of first-passage traces. These metrics turn out to have formulations in terms of the entropies of two different random variables on the set of traces of the processes. In addition, we introduce a measure of `goodness' whereby a trace is termed {\it good} if it satisfies some prescribed linear temporal logic expression. This allows for comparisons of policy processes with respect to the prescribed notion of `goodness'.
Structure-Conditioned Actor-Critic Branches for Quality-Diversity Reinforcement Learning
School of Artificial Intelligence, Nanjing University of Information Science and Technology(南京信息工程大学人工智能学院);Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies, Institute of Cyberspace Security, School of Computer Science and Technology, Harbin Institute of Technology(哈尔滨工业大学计算机科学与技术学院网络空间安全研究院广东省新型安全智能技术重点实验室)
Neural Multi-Objective Combinatorial Optimization with Diversity Enhancement
Sun Yat-sen University(中山大学);Singapore Management University(新加坡管理大学);Eindhoven University of Technology(埃因霍温理工大学);National University of Singapore(新加坡国立大学)
Department of Electrical Engineering and Computer Science, Pennsylvania State University, PA, USA(电气工程与计算机科学系,宾夕法尼亚州立大学);College of Information Sciences and Technology, Pennsylvania State University, PA, USA(信息科学与技术学院,宾夕法尼亚州立大学)
Skill Learning via Policy Diversity Yields Identifiable Representations for Reinforcement Learning
Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);ELLIS Institute Tübingen(图宾根ELLIS研究所);University of Tübingen(图宾根大学);Tübingen AI Center(图宾根人工智能中心);University of Cambridge(剑桥大学);Princeton University(普林斯顿大学)
本文从可辨识表征学习角度研究互信息技能学习,证明对比后继特征方法可在线性变换意义下可证明地恢复环境真实特征,并在 MuJoCo 与 DeepMind Control 上验证了该结论。
Baseline-Free Policy Optimization for Neural Combinatorial Optimization
Facultad de Ingeniería y Ciencias, Universidad Adolfo Ibáñez(阿道夫·伊瓦涅斯大学工程与科学学院);Dirección de Programas, Investigación y Desarrollo, Armada de Chile(智利海军计划、研究与发展局);Millennium Nucleus for Social Data Science (SODAS)(千禧年社会数据科学核心(SODAS));Millennium Nucleus in Data Science for Plant Resilience (PhytoLearning)(千禧年植物韧性数据科学核心(PhytoLearning))
DyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay
Griffith University, Australia(澳大利亚格里菲斯大学);Fudan University, China(复旦大学);Beijing University of Post(北京邮电大学);Shanghai Innovation Institute, China(上海创新研究院);Shanghai Academy of Artificial Intelligence for Science, China(上海人工智能科学研究院)
Diversifying Policy Behaviors with Extrinsic Behavioral Curiosity
School of Data Science, The Chinese University of Hong Kong, Shenzhen, China(数据科学学院,香港中文大学(深圳));CFAR, Agency for Science, Technology and Research, Singapore(CFAR,新加坡科技研究局);College of Computing and Data Science, Nanyang Technological University (NTU), Singapore(计算与数据科学学院,南洋理工大学(NTU))
Constructive Conflict-Driven Multi-Agent Reinforcement Learning for Strategic Diversity
School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);CRISE, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences(复杂系统认知与决策智能重点实验室,中国科学院自动化研究所)
TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion Alignment
Huazhong University of Science and Technology(华中科技大学);Kuaishou Technology(快手科技);Nanyang Technological University(南洋理工大学);Wuhan University(武汉大学);Tsinghua University(清华大学)