Ian Abraham
@i-abr.bsky.social
Asst. Prof @YaleSEAS ME/CS | Prev. 🤖Robotics @CMU_Robotics @NorthwesternU | Lab https://ialab.yale.edu/
My lab and I will be heading to #RSS2025 . My students @hrsathya.bsky.social , @haoxiang-you.bsky.social , Yilang Liu, and Christian Hughes will be around to talk about contact-aware information theory for robot learning and efficient robot learning and control!
Robots 🤖 make and break contact all the time. So what quantifies usefulness of contacts to improve a robot’s understanding of its environment? Check out our latest #RSS2025 paper on contact aware fisher information maximization!! 🧵below—-
How much information is in contact interactions and how can we optimize for them to improve robot learning?
Check out our latest work on Behavior Synthesis via Contact-Aware Fisher Information Maximization led by myself and @i-abr.bsky.social at #RSS2025!
Paper: arxiv.org/pdf/2505.12214
Check out our latest work on Behavior Synthesis via Contact-Aware Fisher Information Maximization led by myself and @i-abr.bsky.social at #RSS2025!
Paper: arxiv.org/pdf/2505.12214
June 19, 2025 at 2:03 PM
My lab and I will be heading to #RSS2025 . My students @hrsathya.bsky.social , @haoxiang-you.bsky.social , Yilang Liu, and Christian Hughes will be around to talk about contact-aware information theory for robot learning and efficient robot learning and control!
Robots 🤖 make and break contact all the time. So what quantifies usefulness of contacts to improve a robot’s understanding of its environment? Check out our latest #RSS2025 paper on contact aware fisher information maximization!! 🧵below—-
How much information is in contact interactions and how can we optimize for them to improve robot learning?
Check out our latest work on Behavior Synthesis via Contact-Aware Fisher Information Maximization led by myself and @i-abr.bsky.social at #RSS2025!
Paper: arxiv.org/pdf/2505.12214
Check out our latest work on Behavior Synthesis via Contact-Aware Fisher Information Maximization led by myself and @i-abr.bsky.social at #RSS2025!
Paper: arxiv.org/pdf/2505.12214
June 2, 2025 at 9:24 PM
Robots 🤖 make and break contact all the time. So what quantifies usefulness of contacts to improve a robot’s understanding of its environment? Check out our latest #RSS2025 paper on contact aware fisher information maximization!! 🧵below—-
🚀🚀🚀 Check out this work out of my lab on accelerating visual-policy learning by exploiting computational structures facilitated by diff sims!! 🧵 below on why its exciting!!!
🚀 Excited to introduce d.va — a computationally efficient approach for learning visual policies!
⚡ Trains in minutes on a laptop and consistently beats existing baselines in reward.
This work was done in collaboration with Yilang Liu and Ian Abraham.
⚡ Trains in minutes on a laptop and consistently beats existing baselines in reward.
This work was done in collaboration with Yilang Liu and Ian Abraham.
May 29, 2025 at 12:14 PM
🚀🚀🚀 Check out this work out of my lab on accelerating visual-policy learning by exploiting computational structures facilitated by diff sims!! 🧵 below on why its exciting!!!
Check out the latest work by my student!!! ~~thread coming soon~~
🚀 Excited to introduce d.va — a computationally efficient approach for learning visual policies!
⚡ Trains in minutes on a laptop and consistently beats existing baselines in reward.
This work was done in collaboration with Yilang Liu and Ian Abraham.
⚡ Trains in minutes on a laptop and consistently beats existing baselines in reward.
This work was done in collaboration with Yilang Liu and Ian Abraham.
May 23, 2025 at 2:55 PM
Check out the latest work by my student!!! ~~thread coming soon~~
My student Hrishi and I will be presenting our work titled "Behavior Synthesis via Contact-Aware Fisher Information Maximization" @roboticsscisys.bsky.social where we investigate contact-based behavior that aids in physical parameter learning! More info to come --- stay tuned
📢 Check out the list of accepted papers for Robotics: Science and Systems 2025: roboticsconference.org/program/pape...
📜 You can also find the complete conference program here: roboticsconference.org/program/over...
🌴 We look forward to seeing you at #RSS2025 in Los Angeles June 21st-25th!
📜 You can also find the complete conference program here: roboticsconference.org/program/over...
🌴 We look forward to seeing you at #RSS2025 in Los Angeles June 21st-25th!
Accepted Papers | Robotics: Science and Systems
RSS 2025, June 21, 2025 - June 25, 2025, Los Angeles, California
roboticsconference.org
May 17, 2025 at 12:42 PM
My student Hrishi and I will be presenting our work titled "Behavior Synthesis via Contact-Aware Fisher Information Maximization" @roboticsscisys.bsky.social where we investigate contact-based behavior that aids in physical parameter learning! More info to come --- stay tuned
My lab and I will be @ICRA in Atlanta showcasing three papers on ergodic search!
lnkd.in/eh3NKwdh
lnkd.in/eqQRJnwS
lnkd.in/e62mR8X4
more details below in comments!
lnkd.in/eh3NKwdh
lnkd.in/eqQRJnwS
lnkd.in/e62mR8X4
more details below in comments!
May 15, 2025 at 10:08 PM
My lab and I will be @ICRA in Atlanta showcasing three papers on ergodic search!
lnkd.in/eh3NKwdh
lnkd.in/eqQRJnwS
lnkd.in/e62mR8X4
more details below in comments!
lnkd.in/eh3NKwdh
lnkd.in/eqQRJnwS
lnkd.in/e62mR8X4
more details below in comments!
We are excited to announce the 2025 ICRA Workshop on Structured Learning for Efficient, Reliable and Transparent Robots (SRL) happening in Atlanta, GA May 23rd!
Workshop paper submission link: docs.google.com/forms/d/e/1F...
Website: sites.google.com/view/srl-icr...
Workshop paper submission link: docs.google.com/forms/d/e/1F...
Website: sites.google.com/view/srl-icr...
March 7, 2025 at 1:04 PM
We are excited to announce the 2025 ICRA Workshop on Structured Learning for Efficient, Reliable and Transparent Robots (SRL) happening in Atlanta, GA May 23rd!
Workshop paper submission link: docs.google.com/forms/d/e/1F...
Website: sites.google.com/view/srl-icr...
Workshop paper submission link: docs.google.com/forms/d/e/1F...
Website: sites.google.com/view/srl-icr...
Reposted by Ian Abraham
"You want to stop, but the amount of time and energy you’ve put into this makes it hard to quit."
Elden Ring or Tenure-Track Professor?
1. You’ve invested an enormous amount of time and energy, and you still have no idea what is happening. 2. Everybody you meet speaks in half riddle...
buff.ly
February 15, 2025 at 9:09 PM
"You want to stop, but the amount of time and energy you’ve put into this makes it hard to quit."
Reposted by Ian Abraham
We don't want vision-language-action models to remain bounded by human performance, but training these large models with reinforcement learning is difficult. How can we handle this? Short blog post: open.substack.com/pub/itcanthi...
Paper notes: Improving Vision-Language-Action Model with Online Reinforcement Learning
Many vision-language-action models really don’t work that well.
open.substack.com
February 6, 2025 at 2:39 PM
We don't want vision-language-action models to remain bounded by human performance, but training these large models with reinforcement learning is difficult. How can we handle this? Short blog post: open.substack.com/pub/itcanthi...
Reposted by Ian Abraham
So. Uh. Well. Uhm.
www.youtube.com/watch?v=X2Ux...
www.youtube.com/watch?v=X2Ux...
Unitree B2-W Talent Awakening! 🥳
YouTube video by Unitree Robotics
www.youtube.com
December 23, 2024 at 6:02 PM
So. Uh. Well. Uhm.
www.youtube.com/watch?v=X2Ux...
www.youtube.com/watch?v=X2Ux...
Reposted by Ian Abraham
I am attending this week ISRR in Long Beach, CA, where I presented a survey/position paper on "The State of Robot Motion Generation". It reviews techniques developed for robot motion generation across many sub-disciplines and recommends research directions forward.
Paper: arxiv.org/abs/2410.12172
Paper: arxiv.org/abs/2410.12172
December 10, 2024 at 8:19 PM
I am attending this week ISRR in Long Beach, CA, where I presented a survey/position paper on "The State of Robot Motion Generation". It reviews techniques developed for robot motion generation across many sub-disciplines and recommends research directions forward.
Paper: arxiv.org/abs/2410.12172
Paper: arxiv.org/abs/2410.12172