Designing robots capable of generating interpretable behavior is a prerequisite for achieving effective human-robot collaboration. This means that the robots need to be capable of generating behavior that aligns with human expectations and, when required, provide explanations to the humans in the loop. However, exhibiting such behavior in arbitrary environments could be quite expensive for robots, and in some cases, the robot may not even be able to exhibit the expected behavior. Given structured environments (like warehouses and restaurants), it may be possible to design the environment so as to boost the interpretability of the robot's behavior or to shape the human's expectations of the robot's behavior. In this paper, we investigate the opportunities and limitations of environment design as a tool to promote a type of interpretable behavior -- known in the literature as explicable behavior. We formulate a novel environment design framework that considers design over multiple tasks and over a time horizon. In addition, we explore the longitudinal aspect of explicable behavior and the trade-off that arises between the cost of design and the cost of generating explicable behavior over a time horizon.
Designing explainable artificial intelligence with active inference: A framework for transparent introspection and decision-making
VERSES Research Lab（VERSES研究实验室）；Université du Québec à Montréal（魁北克大学蒙特利尔分校）；Humboldt-Universität zu Berlin（柏林洪堡大学）；Macquarie University（麦考瑞大学）；University College London（伦敦大学学院）
Designing Interpretable ML System to Enhance Trust in Healthcare: A Systematic Review to Proposed Responsible Clinician-AI-Collaboration Framework
Grado Department of Industrial & Systems Engineering, Virginia Tech（弗吉尼亚理工大学格拉多工业与系统工程系）；Institute for Intelligent Systems Research and Innovation (IISRI), Deakin University（迪肯大学智能系统研究与创新研究所）；School of Mathematics, Physics, and Computing, University of Southern Queensland（南昆士兰大学数学、物理与计算学院）；Center for Health Research, University of Southern Queensland（南昆士兰大学健康研究中心）
BuildEvo: Designing Building Energy Consumption Forecasting Heuristics via LLM-driven Evolution
Department of the Built Environment, National University of Singapore（新加坡国立大学建筑环境系）；Department of Industrial and Systems Engineering, Korea Advanced Institute of Science and Technology（韩国科学技术院工业与系统工程系）
AutoDIME: Automatic Design of Interesting Multi-Agent Environments
OpenAI（开放人工智能研究中心）
设计一个RL代理可以学习有趣和有用的技能的环境分布是一个具有挑战性和理解不足的任务，对于多代理环境，困难只会加剧。一种方法是训练第二个RL代理，称为教师，谁的样本环境，有利于学生代理的学习。然而，大多数以前的建议，教师奖励不直接推广到多智能体设置。我们研究了一组来自预测问题的内在教师奖励，这些问题可以应用于多智能体设置，并在Mujoco任务中对其进行评估，例如多智能体Hide and Seek以及诊断单智能体迷宫任务。在所考虑的内在奖励中，我们发现价值分歧在不同任务中最为一致，导致在捉迷藏和迷宫任务中更快、更可靠地出现高级技能。考虑的另一个候选内在奖励，价值预测误差，在捉迷藏中也很好，但在随机环境中容易受到噪音电视风格的干扰。政策分歧在迷宫任务中表现良好，但在捉迷藏中没有加快学习。我们的研究结果表明，内在的教师奖励，特别是价值分歧，是一个很有前途的方法，自动化的单和多代理环境设计。
BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments
Stanford University（斯坦福大学）；Arc Institute（Arc研究所）；Gladstone-UCSF Institute of Genomic Immunology（Gladstone-加州大学旧金山分校基因组免疫学研究所）；University of California, San Francisco（加州大学旧金山分校）；UCSF Helen Diller Family Comprehensive Cancer Center（加州大学旧金山分校海伦·迪勒家族综合癌症中心）
