Reinforcement learning isn't just for robots! In the summer before starting my CS Ph.D. at
#Stanford
, I spent three amazing months with the U.S.
#Navy
Marine Mammal program, where I got to observe how animals learn from reinforcement! (1/3)
Super excited to share our work on using audio to help with visually occluded tasks, like extracting keys from a paper bag! We also show that the robot is able to distinguish between two materials by their sounds. Feel free to check out our paper and project website.
#RSS2022
Can robots deal with occlusion?
We put a microphone on a robot's gripper & found that audio helps robots learn to solve tasks amidst occlusion.
#RSS2022
paper:
w/
@du_maximilian
, Olivia Lee,
@SurajNair_1
🧵(1/4)
I'm excited to announce the release of my audio story, "Orca Boy" through the Stanford Storytelling Project! Listen to it anywhere you get your podcasts by searching "State of the Human"
#killerwhaletrainer
#seaworld
@stanfordstory
Here’s an article I wrote about why we shouldn’t romanticize the intelligence of non-human animals. As an AI researcher, I believe it is critical to see intelligence as multifaceted and context-specific
@ZCCZHANG
@SurajNair_1
@DorsaSadigh
@chelseabfinn
Hi! We anticipate that retrieval can benefit long-horizon tasks because we retrieve individual transitions, agnostic to horizon length. In our appendix, we try a similar experiment, where we show that retrieval can help us stitch individual tasks together. (Appendix D.A)