Across the country, parents are putting their children to bed, reading stories, singing lullabies—and in the back of their minds, they’re worried about what might happen tomorrow after they drop their kids off at school, or take them to a grocery store or any other public space.
Excited to talk about Adversarial Intrinsic Motivation (AIM), accepted to NeurIPS 2021. It learns an intrinsic reward signal that guides the agent's exploration to find the goal in goal-conditioned reinforcement learning. See an example of the system in action below ... 1/🧵
Congrats to UT Austin Villa on another successful year at #RoboCup! 🤖⚽️
The 3 teams earned:
• 1st in 3D soccer sim league (9th time in 10 years!)
• 3rd in RoboCup@Home
• 4th in Standard Platform League
More details at: https://t.co/WJqlKhPcrP
I'll be presenting my work with @mauriciogtec@scottniekum & @PeterStone_TX titled "Adversarial Intrinsic Motivation for Reinforcement Learning" at the ICRA Workshop Towards Curious Robots tomorrow (4th June). Drop by if you can!
https://t.co/vGvyGWMdUp