Jeff Clune: Genetic Algorithms, Quality-Diversity, Curiosity

In episode 40 of The Gradient Podcast, Andrey Kurenkov speaks to Professor Jeff Clune.

Jeff is an Associate Professor of Computer Science at the University of British Columbia and a Faculty Member of the Vector Institute. Previously, he was a Research Team Leader at OpenAI and before that a Senior Research Manager and founding member of Uber AI Labs, and prior to that he was an Associate Professor in Computer Science at the University of Wyoming.

Subscribe to The Gradient Podcast:  Apple Podcasts  | Spotify | Pocket Casts | RSS
Follow The Gradient on Twitter

Outline:

• (00:00) Intro

• (01:05) Path into AI

• (08:05) Studying biology with simulations

• (10:30) Overview of genetic algorithms

• (14:00) Evolving gaits with genetic algorithms

  • Evolving coordinated quadruped gaits with the HyperNEAT generative encoding

  • Evolving Robot Gaits in Hardware: the HyperNEAT Generative Encoding Vs.Parameter Optimization   

  • Aracna: An open-source quadruped platform for evolutionary robotics

• (20:00) Quality-Diversity Algorithms

  • Robots that can adapt like animals 

  • Illuminating search spaces by mapping elites

• (27:00) Evolving Soft Robots

  • Unshackling Evolution: Evolving Soft Robots with Multiple Materials and a Powerful Generative Encoding

  • Video: Evolving Soft Robots with Multiple Materials (muscle, bone, etc.)

• (32:15) Genetic algorithms for studying Evolution

  • The Evolutionary Origins of Modularity

  • The evolutionary origins of hierarchy

• (39:30) Modularity for Catastrophic Forgetting

  • Neural Modularity Helps Organisms Evolve to Learn New Skills without Forgetting Old Skills

• (45:15) Curiosity for Learning Diverse Skills

  • Curiosity search: producing generalists by encouraging individuals to continually explore and acquire skills throughout their lifetime

  • Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents

  • Go-Explore: a New Approach for Hard-Exploration Problems

• (51:15) Evolving Environments 

  • AI-GAs: AI-generating algorithms, an alternate paradigm for producing general artificial intelligence

  • Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions

  • Enhanced POET: Open-Ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their Solutions

• (58:3) The Surprising Creativity of Digital Evolution

  • The surprising creativity of digital evolution: A collection of anecdotes from the evolutionary computation and artificial life research communities

• (1:04:28) Hobbies Outside of Research

• (1:07:25) Outro