Input devices, such as buttons and sliders, are the foundation of any interface. The typical user-centered design workflow requires the developers and users to go through many iterations of design, implementation, and analysis. The procedure is inefficient, and human decisions highly bias the results. While computational methods are used to assist various design tasks, there has not been any holistic approach to automate the design of input components. My thesis proposed a series of Computational Input Design workflows: I envision a sample-efficient multi-objective optimization algorithm that cleverly selects design instances, which are instantly deployed on physical simulators. A meta-reinforcement learning user model then simulates the user behaviors when using the design instance upon the simulators. The new workflows derive Pareto-optimal designs with high efficiency and automation. I demonstrate designing a push-button via the proposed methods. The resulting designs outperform the known baselines. The Computational Input Design process can be generalized to other devices, such as joystick, touchscreen, mouse, controller, etc.
"ChatGPT, what am I missing?": Designing AI Workflows around Professional Task Structure to Shape Analytic AI Use
AI Institute, Harvard Business School(哈佛商学院人工智能研究所);Frontline Associates(前线协会);Emory University School of Law(埃默里大学法学院);Laboratory for Innovation Science at Harvard(哈佛大学创新科学实验室);Harvard John A. Paulson School of Engineering and Applied Sciences(哈佛约翰·A·保尔森工程与应用科学学院)
Developer Insights into Designing AI-Based Computer Perception Tools
Center for Ethics and Health Policy, Baylor College of Medicine(贝勒医学院伦理与健康政策中心);Department of Psychiatry and Behavioral Sciences, Baylor College of Medicine(贝勒医学院精神病学与行为科学系);Center for Bioethics, Harvard Medical School(哈佛医学院生命伦理学中心)
AIAP: A No-Code Workflow Builder for Non-Experts with Natural Language and Multi-Agent Collaboration
Department of Animation, \ University of Art \& Design Uiwang-si Gyeonggi-do Republic of Korea;Graduate School of Information, \ University Seoul Republic of Korea;Department of Animation, \ University of Art \& Design;Graduate School of Information, \ University
A Collective Knowledge workflow for collaborative research into multi-objective autotuning and machine learning techniques
开发高效的软件和硬件从未如此困难,无论是小型物联网设备还是Exascale超级计算机。除了不断增长的设计和优化复杂性,还存在更根本的问题,如缺乏有效的软件/硬件协同设计所需的跨学科知识,以及学术界和工业界之间日益扩大的技术转让差距。我们介绍我们的新的教育举措,以解决这些问题,通过开发集体知识(CK),一个统一的实验框架计算机系统的研究和开发。我们使用CK来教社区如何使他们的研究工件和实验工作流程便携,可复制,可定制和可重复使用,同时实现可持续的研发和促进技术转让。我们还演示了如何重新设计多目标自动调整和机器学习作为一个便携式和可扩展的CK工作流程。这样的工作流程使研究人员能够使用不同的应用程序,数据集和工具进行实验;跨不同平台进行众包实验;共享实验结果,模型,可视化;使用简单的JSON API逐渐暴露更多的设计和优化选择;并最终建立在彼此的发现之上。作为第一个实际步骤,我们已经实现了可定制的编译器自动调优,在Raspberry Pi 3设备上对各种工作负载进行众包优化,将执行时间和代码大小减少了40%,并应用机器学习来预测优化。我们希望这种方法将有助于教学生如何建立在彼此的工作,使有效的和自我优化的软件/硬件/模型堆栈的新兴工作负载。
Vibe Coding in Product Teams: Reconfiguring AI-Assisted Workflows, Prototyping, and Collaboration
Netherlands and MIT Media Lab(荷兰和MIT媒体实验室);Notion Labs(Notion实验室);Collov AI;Cornell University(康奈尔大学);Goizueta Business School, Emory University(埃默里大学戈伊兹埃塔商学院);Centrum Wiskunde & Informatica, The Netherlands and Utrecht University, The Netherlands(荷兰代尔夫特理工大学和乌得勒支大学)