澳门澳门威尼娱乐场 http://huaanyiliao.cn:443 The SMARTech digital repository system captures, stores, indexes, preserves, and distributes digital research material. Sat, 24 Apr 2021 17:39:05 GMT 2021-04-24T17:39:05Z 澳门澳门威尼娱乐场 http://hdl.handle.net/1853/64460 Design Justice: Community-Led Practices to Build the Worlds We Need Costanza-Chock, Sasha Dr. Sasha Costanza-Chock presents an overview of their book Design Justice: Community-Led Practices to Build the Worlds We Need, published by the MIT Press in 2020. The book is an exploration of how we might re-imagine design to be led by marginalized communities as a tool to help dismantle structural inequality, advance collective liberation, and support ecological survival. In this book talk, Dr. Costanza-Chock presents an overview of key themes, concepts, and excerpts from the text, followed by a discussion with the audience. Presented online April 6, 2021, 5:00 p.m.-6:00 p.m.; Sasha Costanza-Chock (they/them or she/her) is a researcher and designer who works to support community-led processes that build shared power, move towards collective liberation, and advance ecological survival. They are known for their work on networked social movements, transformative media organizing, and design justice. Sasha is a Research Scientist at the Massachusetts Institute of Technology with a joint appointment in Media Arts & Sciences at the MIT Media Lab and the Department of Urban Studies+Planning. They are a Senior Research Fellow at the Algorithmic Justice League and a Faculty Affiliate with the Berkman-Klein Center for Internet & Society at Harvard University. Sasha is a board member of Allied Media Projects and a member of the Steering Committee of the Design Justice Network. Sasha is the author of two books and numerous journal articles, book chapters, and other research publications. Their new book, Design Justice: Community-Led Practices to Build the Worlds We Need, was published by the MIT Press in 2020.; Runtime: 54:12 minutes Tue, 06 Apr 2021 00:00:00 GMT http://hdl.handle.net/1853/64460 2021-04-06T00:00:00Z Costanza-Chock, Sasha Dr. Sasha Costanza-Chock presents an overview of their book Design Justice: Community-Led Practices to Build the Worlds We Need, published by the MIT Press in 2020. The book is an exploration of how we might re-imagine design to be led by marginalized communities as a tool to help dismantle structural inequality, advance collective liberation, and support ecological survival. In this book talk, Dr. Costanza-Chock presents an overview of key themes, concepts, and excerpts from the text, followed by a discussion with the audience. 澳门澳门威尼娱乐场 http://hdl.handle.net/1853/64459 Curator Conversation: Hin Bredendieck: From Aurich to Atlanta Henderson, Kirk; Stamm, Rainer; Budd, James (Jim) G. A discussion between Professor Dr. Rainer Stamm, director of the Landesmuseum Oldenburg (Oldenburg State Museum for Art and Cultural History), and Kirk Henderson, Exhibits Program Manager for the Georgia Tech Library, moderated by Jim Budd, chair of the School of Industrial Design. This event will start with a virtual tour through the exhibit, followed by a discussion between the exhibit’s curators. Budd, Stamm, and Henderson will discuss their cross-continental collaboration that led to the exhibit, as well as the curatorial themes that emerged through their work together. Presented online April 8, 2021, 11:00 a.m.-11:59 a.m.; From Aurich to Atlanta, a new exhibition at the Georgia Tech Library, showcases the life and work of Bauhaus-educated designer Hin Bredendieck. A 1930 graduate of the famed Bauhaus School of Design in Germany, Bredendieck was a student and colleague of design luminaries such as Walter Gropius, Paul Klee, and Laslo Moholy-Nagy. Bredendieck emigrated to the United States in 1937 to escape the political turmoil of pre-war Germany. With his fellow Bauhaus emigres, he brought the Bauhaus design sense and educational methods to America, teaching first at the new Institute of Design in Chicago, and later founding the department of industrial design at Georgia Tech.; Kirk Henderson is the Exhibitions Program Manager for the Georgia Tech Library. Henderson coordinates the design and curation of exhibits in the Library, many of which feature the Archives’ Special Collections. Formerly a curator at the Atlanta History Center, he collaborated on major exhibit installations about regional history and the American Civil War.; Dr. Rainer Stamm, co-author of Hin Bredendieck: From Aurich to Atlanta, is director of the Oldenburg State Museum for Art and Cultural History. He is an honorary professor of art history at the University of Bremen, with a special focus on modern art history, museum history, the history of photography and art market history of the early 20th century.; James (Jim) G. Budd is the Chair, School of Industrial Design, Georgia Institute of Technology, Atlanta. GA.; Runtime: 56:53 minutes Thu, 08 Apr 2021 00:00:00 GMT http://hdl.handle.net/1853/64459 2021-04-08T00:00:00Z Henderson, Kirk Stamm, Rainer Budd, James (Jim) G. A discussion between Professor Dr. Rainer Stamm, director of the Landesmuseum Oldenburg (Oldenburg State Museum for Art and Cultural History), and Kirk Henderson, Exhibits Program Manager for the Georgia Tech Library, moderated by Jim Budd, chair of the School of Industrial Design. This event will start with a virtual tour through the exhibit, followed by a discussion between the exhibit’s curators. Budd, Stamm, and Henderson will discuss their cross-continental collaboration that led to the exhibit, as well as the curatorial themes that emerged through their work together. 澳门澳门威尼娱乐场 http://hdl.handle.net/1853/64458 Conflict-Aware Risk-Averse and Safe Reinforcement Learning: A Meta-Cognitive Learning Framework Modares, Hamidreza While the success of reinforcement learning (RL) in computer games has shown impressive engineering feat, unlike the computer games, safety-critical settings such as unmanned vehicles must thrash around in the real world, which makes the entire enterprise unpredictable. Standard RL practice generally implants pre-specified performance metrics or objectives into the RL agent to encode the designers’ intention and preferences in achieving different and sometimes conflicting goals (e.g., cost efficiency, safety, speed of response, accuracy, etc.). Optimizing pre-specified performance metrics, however, cannot provide safety and performance guarantees across a vast variety of circumstances that the system might encounter in non-stationary and hostile environments. In this talk, I will discuss novel metacognitive RL algorithms to learn not only a control policy that optimizes accumulated reward values, but also what reward functions to optimize in the first place to formally assure safety with a good enough performance. I will present safe RL algorithms that adapt the focus of attention of RL algorithm to its variety of performance and safety objectives to resolve conflict and thus assure the feasibility of the reward function in a new circumstance. Moreover, model-free RL algorithms will be presented to solve the risk-averse optimal control (RAOC) problem to optimize the expected utility of outcomes while reducing the variance of cost under aleatory uncertainties (i.e., randomness). This is because, performance-critical systems must not only optimize the expected performance, but also reduce its variance to avoid performance fluctuation during RL’s course of operation. To solve the RAOC problem, I will present the three variants of RL algorithms and analyze their advantages and preferences for different situations/systems: 1) a one-shot static convex program based RL, 2) an iterative value iteration algorithm that solves a linear programming optimization at each iteration, and 3) an iterative policy iteration algorithm that solves a convex optimization at each iteration and guarantees the stability of the consecutive control policies. Presented online April 14, 2021 at 12:15 p.m.; Hamidreza Modares is an Assistant Professor in the Department of Mechanical Engineering at Michigan State University. Prior to joining Michigan State University, he was an Assistant Professor in the Department of Electrical Engineering, Missouri University of Science and Technology. His current research interests include control and security of cyber–physical systems, machine learning in control, distributed control of multi-agent systems, and robotics. He is an Associate Editor of IEEE Transactions on Neural Networks and Learning Systems.; Runtime: 60:34 minutes Wed, 14 Apr 2021 00:00:00 GMT http://hdl.handle.net/1853/64458 2021-04-14T00:00:00Z Modares, Hamidreza While the success of reinforcement learning (RL) in computer games has shown impressive engineering feat, unlike the computer games, safety-critical settings such as unmanned vehicles must thrash around in the real world, which makes the entire enterprise unpredictable. Standard RL practice generally implants pre-specified performance metrics or objectives into the RL agent to encode the designers’ intention and preferences in achieving different and sometimes conflicting goals (e.g., cost efficiency, safety, speed of response, accuracy, etc.). Optimizing pre-specified performance metrics, however, cannot provide safety and performance guarantees across a vast variety of circumstances that the system might encounter in non-stationary and hostile environments. In this talk, I will discuss novel metacognitive RL algorithms to learn not only a control policy that optimizes accumulated reward values, but also what reward functions to optimize in the first place to formally assure safety with a good enough performance. I will present safe RL algorithms that adapt the focus of attention of RL algorithm to its variety of performance and safety objectives to resolve conflict and thus assure the feasibility of the reward function in a new circumstance. Moreover, model-free RL algorithms will be presented to solve the risk-averse optimal control (RAOC) problem to optimize the expected utility of outcomes while reducing the variance of cost under aleatory uncertainties (i.e., randomness). This is because, performance-critical systems must not only optimize the expected performance, but also reduce its variance to avoid performance fluctuation during RL’s course of operation. To solve the RAOC problem, I will present the three variants of RL algorithms and analyze their advantages and preferences for different situations/systems: 1) a one-shot static convex program based RL, 2) an iterative value iteration algorithm that solves a linear programming optimization at each iteration, and 3) an iterative policy iteration algorithm that solves a convex optimization at each iteration and guarantees the stability of the consecutive control policies. 澳门澳门威尼娱乐场 http://hdl.handle.net/1853/64457 Georgia Tech PTRC Annual Activities Report, 2020-2021 Li, Lisha Mon, 01 Mar 2021 00:00:00 GMT http://hdl.handle.net/1853/64457 2021-03-01T00:00:00Z Li, Lisha