GECCO2021 — Keynote — Joshua Tenenbaum — Reverse-engineering core common sense with the tools [...]
Reverse-engineering core common sense with the tools of probabilistic programs, game-style simulation engines, and inductive program synthesis
Joshua Tenenbaum (MIT, USA)
None of today’s AI systems or approaches comes anywhere close to capturing the common sense of a toddler, or even a 3-month old infant. I will talk about some of the challenges facing conventional machine learning paradigms, such as end-to-end unsupervised learning in deep networks and deep reinforcement learning, and discuss some initial, small steps we have taken with an alternative cognitively-inspired AI approach. This requires us to develop a different engineering toolset, based on probabilistic programs, game-style simulation programs as general-purpose startup software (or “the game engine in the head”), and learning as programming (or “the child as hacker”).
Joshua Tenenbaum is the Paul E. Newton Career Development Professor of Cognitive Science and Computation in the Department of Brain and Cognitive Sciences, and a member of the Computer Science and Artificial Intelligence Laboratory at MIT. He received his Ph.D. from MIT in 1999 and after a brief postdoc with the MIT AI Lab, he joined the Stanford University faculty as Assistant Professor of Psychology and (by courtesy) Computer Science. He returned to MIT as a faculty member in 2002. He currently serves as Associate Editor of the journal Cognitive Science, and he has been active on the program committees of the Neural Information Processing Systems (NIPS) and Cognitive Science (CogSci) conferences. In 2019, he was named a MacArthur Fellow.
GECCO 2021
The Genetic and Evolutionary Computation Conference
July 10-14, 2021 — Lille, France (online)
https://gecco-2021.sigevo.org
Видео GECCO2021 — Keynote — Joshua Tenenbaum — Reverse-engineering core common sense with the tools [...] канала Association for Computing Machinery (ACM)
Joshua Tenenbaum (MIT, USA)
None of today’s AI systems or approaches comes anywhere close to capturing the common sense of a toddler, or even a 3-month old infant. I will talk about some of the challenges facing conventional machine learning paradigms, such as end-to-end unsupervised learning in deep networks and deep reinforcement learning, and discuss some initial, small steps we have taken with an alternative cognitively-inspired AI approach. This requires us to develop a different engineering toolset, based on probabilistic programs, game-style simulation programs as general-purpose startup software (or “the game engine in the head”), and learning as programming (or “the child as hacker”).
Joshua Tenenbaum is the Paul E. Newton Career Development Professor of Cognitive Science and Computation in the Department of Brain and Cognitive Sciences, and a member of the Computer Science and Artificial Intelligence Laboratory at MIT. He received his Ph.D. from MIT in 1999 and after a brief postdoc with the MIT AI Lab, he joined the Stanford University faculty as Assistant Professor of Psychology and (by courtesy) Computer Science. He returned to MIT as a faculty member in 2002. He currently serves as Associate Editor of the journal Cognitive Science, and he has been active on the program committees of the Neural Information Processing Systems (NIPS) and Cognitive Science (CogSci) conferences. In 2019, he was named a MacArthur Fellow.
GECCO 2021
The Genetic and Evolutionary Computation Conference
July 10-14, 2021 — Lille, France (online)
https://gecco-2021.sigevo.org
Видео GECCO2021 — Keynote — Joshua Tenenbaum — Reverse-engineering core common sense with the tools [...] канала Association for Computing Machinery (ACM)
Показать
Комментарии отсутствуют
Информация о видео
15 июля 2021 г. 19:49:15
01:18:54
Другие видео канала
CACM Dec. 2018 - Point Counterpoint on AI RegulationWelcome ACM Associate EditorsCACM Mar. 2017 - Computational Support for Academic Peer ReviewCACM Mar. 2016 - Hopes, Fears, and Software ObfuscationApril 2022 CACM: Digital Twins and Artificial IntelligenceKDD 2020: Lecture Style Tutorials: Deep Graph Learning Foundations, Advances and ApplicationsVisual Data Analysis: How? When? Why?System Methodology—Holistic Performance Analysis on Modern SystemsKDD 2020: Hands-onTutorials: Deep Learning for Search and Recommender Systems in Practice-Part 5CACM Nov. 2019 - DeepXplore: Automated Whitebox Testing of Deep Learning Systems"Fairness in Machine Learning" with Tulsee DoshiAndrew File System Developers Receive 2016 ACM Software System AwardSeptember 2021 CACM: The Future Is Big GraphsLearning from Data: The Two CulturesMarch 2022 CACM: Automating Data ScienceCACM August 2016 - Ur/Web: A Simple Model for Programming the WebCACM July 2017 - Reimagining the Avatar Dream: Modeling Social Identity in Digital MediaReForm: Integrating Physical and Digital Design through Bidirectional FabricationRunning Linux Apps on Windows How and Why?ACM Software Systems Award 2012CACM July 2018 - Making Machine Learning Robust Against Adversarial Inputs