Long-term Real World Autonomy for Assistive Robots
Abstract: For long-term deployment in real-world environments, robots need to continually learn new concepts to adapt to their ever-changing environments. However, continual learning in real-world environments poses various challenges. First, machine learning (ML) models for continual learning can forget previously learned knowledge when learning new information, a problem called catastrophic forgetting. Second, the only source of supervision for the robots in everyday environments is their non-expert users, who might be unwilling to provide a large number of "perfect" training examples to the robot. In this talk, I will discuss how we can draw inspiration from theories of learning in the human brain to develop ML models that can continually learn from limited data while mitigating catastrophic forgetting. I will then present how these models can be integrated with autonomous robots to learn from non-expert human users. Finally, I will present how these models can be integrated into complete systems that can allow robots to adapt without human supervision while performing assistive tasks in a household-type environment.
Speaker Bio: Dr. Ali Ayub is a Postdoctoral fellow at the University of Waterloo in the Department of Electrical and Computer Engineering, advised by Professor Kerstin Datuenhahn and Professor Chrystopher Nehaniv. He studies lifelong learning for assistive robots that continually learn personalized knowledge from people to assist them in their daily environments. His research combines methods from machine learning and human-robot interaction to develop theoretical frameworks that are integrated into practical systems for human-robot interaction in domains like assistive robotic arms, mobile manipulators, and socially assistive robots. Prior to his postdoc, he earned his Ph.D. and MS from The Pennsylvania State University in Electrical Engineering in 2021 and 2017, respectively. Before joining Penn State, he earned his B.S. in Electrical Engineering at the University of Engineering and Technology, Lahore. He is a recipient of Penn State’s Robert W. Graham Fellowship, the United States Educational Foundation’s Global UGRAD Fellowship, and Google’s diversity, equity, and inclusion (DEI) award. He has also been selected as a Pioneer at the IEEE International Conference on Human-Robot Interaction (HRI).
Видео Long-term Real World Autonomy for Assistive Robots канала WaterlooAI
Speaker Bio: Dr. Ali Ayub is a Postdoctoral fellow at the University of Waterloo in the Department of Electrical and Computer Engineering, advised by Professor Kerstin Datuenhahn and Professor Chrystopher Nehaniv. He studies lifelong learning for assistive robots that continually learn personalized knowledge from people to assist them in their daily environments. His research combines methods from machine learning and human-robot interaction to develop theoretical frameworks that are integrated into practical systems for human-robot interaction in domains like assistive robotic arms, mobile manipulators, and socially assistive robots. Prior to his postdoc, he earned his Ph.D. and MS from The Pennsylvania State University in Electrical Engineering in 2021 and 2017, respectively. Before joining Penn State, he earned his B.S. in Electrical Engineering at the University of Engineering and Technology, Lahore. He is a recipient of Penn State’s Robert W. Graham Fellowship, the United States Educational Foundation’s Global UGRAD Fellowship, and Google’s diversity, equity, and inclusion (DEI) award. He has also been selected as a Pioneer at the IEEE International Conference on Human-Robot Interaction (HRI).
Видео Long-term Real World Autonomy for Assistive Robots канала WaterlooAI
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
The Morality of Artificial Intelligence in WarfareLearning to Execute Prioritized Stacks of Robotic Tasks - Gennaro NotomistaIntegrated Additive Manufacturing and AI Platforms for Smart ManufacturingReinforcement learning in the real world - how to "cheat" and still feel good about it.Safety Assurance of AI-enabled Robotic SystemsWhat is AI? - Undergraduate VideoLet's Talk AI - AI's Role in Navigating Ice-Covered Waters with Zhao PanBayesian Principles for Learning MachinesAI-enabled Knowledge Prediction EngineConditional Generative Adversarial Networks: Iterative Generation and Holistic EvaluationIndustry Day - AI for Supply Chain - Nov 30th, 2021Let's Talk AI - Synthetic Data with Helen ChenKeeping Track of Entities Over Time, Minds, and Knowledge Sources2021 University of Waterloo GRADflix 3rd Place Winner: Ali NasrDeveloping Reinforcement Learning Agents that Learn Many SubtasksLogic-Based Computational Ethics for Autonomous AgentsHigh Performance ManufacturingLet's Talk AI - AI's Future with Gautam KamathFair and Optimal Prediction via Post-Processing - Han ZhaoOptimizing Pre-trained Clinical Embeddings for Automatic COVID-related ICD CodingAI, Conflict, and the Laws of Armed Conflict