In many sequential decision-making tasks, the agent is not able to model the full complexity of the world, which consists of multitudes of relevant and irrelevant information. For example, a person walking along a city street who tries to model all aspects of the world would quickly be overwhelmed by a multitude of shops, cars, and people moving in and out of view, each following their own complex and inscrutable dynamics. Is it possible to turn the agent's firehose of sensory information into a minimal latent state that is both necessary and sufficient for an agent to successfully act in the world? We formulate this question concretely, and propose the Agent Control-Endogenous State Discovery algorithm (AC-State), which has theoretical guarantees and is practically demonstrated to discover the minimal control-endogenous latent state which contains all of the information necessary for controlling the agent, while fully discarding all irrelevant information. This algorithm consists of a multi-step inverse model (predicting actions from distant observations) with an information bottleneck. AC-State enables localization, exploration, and navigation without reward or demonstrations. We demonstrate the discovery of the control-endogenous latent state in three domains: localizing a robot arm with distractions (e.g., changing lighting conditions and background), exploring a maze alongside other agents, and navigating in the Matterport house simulator.
Guaranteed Scalable Learning of Latent Tree Models
University of Maryland(马里兰大学);Microsoft(微软);HEALTH[at]SCALE;Flatiron Health;Georgia Institute of Technology(佐治亚理工学院);California Institute of Technology(加州理工学院)
Causal discovery with endogenous context variables
German Aerospace Center (DLR)(德国航空航天中心 (DLR));Technische Universität Berlin(柏林工业大学);Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI) Dresden / Leipzig(德累斯顿/莱比锡可扩展数据分析与人工智能中心 (ScaDS.AI));Technische Universität Dresden(德累斯顿工业大学)
Diagnosis of Coronavirus Disease 2019 (COVID-19) with Structured Latent Multi-View Representation Learning
College of Intelligence and Computing, Tianjin University(天津大学智能与计算学院);Department of Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology(华中科技大学同济医学院附属同济医院放射科);Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine(上海交通大学医学院附属瑞金医院放射科);Department of Research and Development, Shanghai United Imaging Intelligence Co., Ltd.(上海联影智能医疗科技有限公司研发部);Department of Radiology, China-Japan Union Hospital of Jilin University(吉林大学中日联谊医院放射科)
Department of Computer Science, Emory University(埃默里大学计算机科学系);Department of Computer Science, Virginia Tech(弗吉尼亚理工大学计算机科学系);Department of Electrical and Computer Engineering, Worcester Polytechnic Institute(伍斯特理工学院电气与计算机工程系);Bradley Department of Electrical and Computer Engineering, Virginia Tech(弗吉尼亚理工大学布拉德利电气与计算机工程系)