The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Reinforcement Learning for Industrial AI with Pieter Abbeel - #476
Apr 19, 2021
Pieter Abbeel, a leading Professor at UC Berkeley and Co-founder of Covariant, dives into the cutting-edge world of AI and robotics. He discusses the challenges of transforming AI concepts into practical applications, especially in warehousing. Abbeel highlights the unique blend of unsupervised and reinforcement learning methods that foster curiosity-driven learning. He also unveils his research on pre-trained transformers as versatile computation tools and introduces his new podcast, Robot Brains, focused on bridging AI research with real-world applications.
AI Snips
Chapters
Transcript
Episode notes
AI in Warehousing
- Warehousing and logistics present a significant opportunity for AI robotics due to high demand for automation.
- Existing automation solutions address "legwork," but "handwork" requires more intelligent systems.
Robot Agility
- Adapting robots to specific form factors limits their use.
- Building smarter, more agile robots can unlock their potential for various applications.
Self-Driving Cars vs. Warehouse Robots
- Self-driving car demos exist, but perfect reliability is difficult to achieve like warehouse robots.
- Constrained environments like highways might be an initial focus due to fewer unexpected events.