Pieter Abbeel, Professor at UC Berkeley shares how his Artificial Intelligence lab is using NVIDIA GPUs and deep reinforcement learning to enable a robot to learn on its own.
Abbeel’s robotics work was recently highlighted in Rolling Stone’s story “Inside the Artificial Intelligence Revolution: A Special Report, Pt. 1”.
The video demonstrates how their robot BRETT learns by playing with children’s toys, such as learning how to stack and sort blocks. In just fifteen minutes the robot is able to successfully figure out how to place the block in the matching hole.
For more details about deep reinforcement learning, read the published papers by Peter Abbeel.
Share your GPU-accelerated science with us at http://nvda.ly/Vpjxr and with the world on #ShareYourScience.
Watch more scientists and researchers share how accelerated computing is benefiting their work at http://nvda.ly/X7WpH
Share Your Science: Robots Thinking More Like Humans with Deep Reinforcement Learning
Apr 07, 2016
Discuss (0)

Related resources
- GTC session: Training Highly Dynamic Robots for Complex Tasks in Industrial Applications (Spring 2023)
- GTC session: Teaching Kits for Educators: Priming University Students for the AI and Accelerated Computing Future (Spring 2023)
- GTC session: Generate Synthetic Data Using Omniverse Replicator for Perception Models (Spring 2023)
- Webinar: Isaac Developer Meetup #2 - Build AI-Powered Robots with NVIDIA Isaac Replicator and NVIDIA TAO
- Webinar: NVIDIA Inception: Maximize Your Benefits
- Webinar: Deep Learning Demystified