Go-Explore: A New Type of Algorithm for Hard-exploration Problems Published 2019-01-24 Download video MP4 360p Recommendations 02:45 What is Swarm Intelligence? 17:34 Requirement Specification vs User Stories 19:03 Go-Explore: a New Approach for Hard-Exploration Problems 1:00:15 Ilya Sutskever: OpenAI Meta-Learning and Self-Play | MIT Artificial General Intelligence (AGI) 45:46 Geoffrey Hinton | On working with Ilya, choosing problems, and the power of intuition 10:46 4 Algorithms We Borrowed from Nature 43:32 Symbolic AGI: How the Natural Will Build the Formal 13:43 Swarm intelligence simulation. Project "Screaming Insects" 1:03:55 🚀 TDD, Where Did It All Go Wrong (Ian Cooper) 1:00:06 Joseph Suarez Thesis Defense - Neural MMO 15:22 How Intelligence Evolved | A 600 million year story. 15:28 What are Diffusion Models? 1:15:58 Yann LeCun: Deep Learning, ConvNets, and Self-Supervised Learning | Lex Fridman Podcast #36 19:50 An introduction to Policy Gradient methods - Deep Reinforcement Learning 43:00 The Mastermind Behind GPT-4 and the Future of AI | Ilya Sutskever 18:05 How AI 'Understands' Images (CLIP) - Computerphile 1:35:14 Tutorial on Quality-Diversity Optimisation 1:00:49 The Art of Code - Dylan Beattie 1:05:32 François Chollet - Creating Keras 3 29:59 AI and Machine Learning Solutions for Fighting Financial Crime – Igor Sumkovski, Santander Similar videos 06:18 Deep Learning Summit SF - 'Go-Explore: A New Type of Algorithm for Hard-exploration Problems'. 1:09:07 Go-Explore: Solving Hard Exploration Problems 00:33 Go-Explore scores 21,000 on Pitfall (previous AI record was zero) 33:49 POET: Endlessly Generating Increasingly Complex and Diverse Learning Environments and Solutions 23:26 First Return Then Explore 07:29 How Go-Explore Solved 55 Atari Games 16:01 Reinforcement Learning with sparse rewards 01:52 Montezuma’s Revenge Solved by Go-Explore (Sets Records on Pitfall too) 03:03 "First return, then explore": Atari games solved by Go-Explore 02:11 Go-Explore Minimal Python Implementation 09:06 10 weird algorithms 00:17 Funky Town 1:10:42 Latent State Recovery in Reinforcement Learning - John Langford 18:00 Go-Explore Complex 3D Environments for Automated Reachability Testing 31:22 Learning Exploration Strategies with Meta-Reinforcement Learning 01:35 "First return then explore”: Go-Explore solves all previously unsolved Atari games 15:23 ⚡ Minesweeper Is Hard - Keegan R 3:47:10 David Kipping: Alien Civilizations and Habitable Worlds | Lex Fridman Podcast #355 More results