Showing posts with label learning to human. Show all posts
Showing posts with label learning to human. Show all posts

Thursday, February 22, 2024

How Robots Will Learn to Smell


Parenting a 3-year-old robot
Aug 2023, phys.org

RoboAgent, an artificial intelligence agent that leverages passive observations and active learning to enable a robot to acquire manipulation abilities on par with a toddler. The team's agent learns through a combination of self-experiences and passive observations contained in internet data. As a parent would guide their child, researchers teleoperated the robot through tasks to provide it with useful self-experiences.

Our novel policy architecture allows our agents to reason even with limited experiences, using temporal chunks of movements instead of commonly used per-timestep actions, and learning from videos on the internet, akin to how babies acquire knowledge and behaviors by passively observing their surroundings.

via Carnegie Mellon University and Facebook: RoboAgent and RoboSet Project - Towards Sample Efficient Robot Manipulation with Semantic Augmentations and Action Chunking. Homanga Bharadhwaj et. al. 

Post Script: Partnership between Carnegie Mellon University and Meta, which is exactly how you wanted this to happen.