Abstract: Despite the progress made in AI, especially in the successful deployment of deep learning for many useful tasks, the systems involved typically require a huge number of training instances, and hence a long time for training. As a result, these systems are not able to rapidly adapt to changing rules and constraints in the environment. This is unlike humans, who are usually able to learn with only a handful of experiences. This hampers the deployment of, say, an adaptive robot that can learn and act rapidly in the ever-changing environment of a home, office, factory, or disaster area. Thus, it is necessary for an AI or robotic system to achieve human performance not only in terms of the “level” or “score” (e.g., success rate in classification, score in Atari game playing, etc.) but also in terms of the speed with which the level or score can be achieved. In contrast with earlier DeepMind’s effort on Atari games, in which they demonstrated the ability of a deep reinforcement learning system to learn and play the games at human level in terms of score, we describe a system that is able to learn causal rules rapidly in an Atari game environment and achieve human-like performance in terms of both score and time.
In this talk, attendees will learn about a new technique and direction of research in AI that is just beginning to be explored, but has important future applications. Notably, AI and robotic systems need to learn rapidly in a fast-changing environment, and learning of causality rapidly, is key. Current techniques such as deep learning and deep reinforcement learning require too much data and time for learning for any practical AI and robotic system to be possible. As articulate by Stanford professor Andrew Ng, reinforcement learning currently has more PR value than practical value, as the game environments such as Go or chess on which these algorithms have been successfully tested on are stationary – the rules do not change – but such is not the case in real-world situations. Professor Judea Pearl of UCLA, the 2011 recipient of the Turing Award (the “Nobel Prize” for computer science), had said: “To build truly intelligent machines, teach them cause and effect”. Such is the importance of the relatively new domain of causality and causal learning. These are the main takeaways for the attendees.
Bio: Seng-Beng Ho is currently Senior Scientist & Deputy Director, Department of Social & Cognitive Computing, Institute of High Performance Computing, Agency of Science, Technology & Research, Singapore. He obtained his Ph.D. in Cognitive Science (AI, Neuroscience, Psychology, & Linguistics) and M.Sc. in Computer Science from the University of Wisconsin, Madison, U.S.A. He has a B.E. in Electronic Engineering from the University of Western Australia. He is the author of a monograph published in June 2016 by Springer International entitled “Principles of Noology: Toward a Theory and Science of Intelligence”. In the book, he presents a principled and fundamental theoretical framework that is critical for building truly general AI systems. Prior to the current position, for 11 years he was President of E-Book Systems Pte Ltd, an e-book Technology company he founded with offices in the Silicon Valley, Beijing, Tokyo, Germany, and Singapore. The company developed and marketed a patented, novel 3D page-flipping technology platform for e-book. Prior to that, he lectured and conducted research on AI and Cognitive Science at the Department of Information Systems and Computer Science, National University of Singapore. He holds 36 U.S. and world-wide patents related to e-book technology and has published more than 30 papers in the field of AI since returning from industry.