I've always wondered how to pronounce ∃

Delighted to share our perspective piece “Language is primarily a tool for communication rather than thought” now out in Nature: nature.com/articles/s4158… (with long-term collaborators (and friends) steven t. piantadosi and Ted Gibson, Language Lab MIT) 1/n

A model that honestly says it doesn’t know what will happen is not necessarily a bad model. Even with all the data, many things are hard to predict. We have to stop believing data is enough.

Robots can infer rewards from human feedback, but what about safety boundaries? Our work, Safe MPC Alignment (arxiv.org/abs/2407.04216), led by Zhixian Xie, collab’d with Wenlong Zhang, Yi Ren, Zhaoran Wang, and George Pappas, shows it’s possible and can be human-data efficient!

Jeff Bezos+his ‘shadow’ colleague read on an airplane trip read my 2004 essay ‘The Cognitive Style PowerPoint” where I said no more PP ever, and instead use 4-page essay with sentences read at start meeting. A history of Amazon said my method gave a them tenfold advantage.

The U.S. National Science Foundation announced the Human AugmentatioN via Dexterity (HAND) Engineering Research Center (ERC)! Our Center will revolutionize how #robots augment human labor using dexterous robot hands with AI-powered skills and intuitive interfaces. new.nsf.gov/news/nsf-annou…

The HAND ERC is excited to work with talent on exciting breakthrough technology, democratizing access to robotics. We are actively seeking an administrative director, additional industrial members/partners, as well as collaborators on SECO grants. handerc.org #Haptics

We’re pleased to share HAND’s new overview video. Spread the excitement within your networks. youtube.com/watch?v=1lf36p… #DexterousManipulation #RobotDexterity #TactileRobotics #HANDERC

Northwestern vaulted to the No. 6 spot in the U.S. News & World Report’s annual National Universities rankings, the highest in the school’s history. — Reported by Nineth Kanieski Koso dailynorthwestern.com/2024/09/24/cam…

Richard Feynman’s 1985 lecture on computer science. ✍️

Evaluation in robot learning papers, or, please stop using only success rate a paper and a 🧵 arxiv.org/abs/2409.09491

I am excited for this next challenge. cmu.edu/news/stories/a…

The coolest visual representation of a Fourier transform. 😲 [bit.ly/3uCu4a9]

The embodied AI revolution continues! ✊ You can now become an... Assistant Professor of Embodied AI at Northwestern: mccormick.northwestern.edu/mechanical/car…

Spotted at ETH's Automatic Control Laboratory! 😀

I'm excited to share some of my recent papers. [1/5] In our recent IJRR paper (journals.sagepub.com/doi/abs/10.117…), we propose a scalable game-theoretic social navigation algorithm and deploy it on an untethered quadruped with an Nvidia Jetson. GitHub repo: github.com/MurpheyLab/brne

This is probably the coolest use of robots in entertainment I have seen.

𝗖𝗮𝗻 𝗿𝗼𝗯𝗼𝘁𝘀 𝗲𝘅𝗽𝗹𝗼𝗿𝗲 𝘀𝗽𝗮𝘁𝗶𝗮𝗹 𝗱𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻𝘀 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲 𝘄𝗮𝘆 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗺𝗼𝗱𝗲𝗹𝘀 𝘀𝗮𝗺𝗽𝗹𝗲 𝗳𝗿𝗼𝗺 𝘁𝗵𝗲𝗺? I’m excited to share our #RSS2025 paper "Flow Matching Ergodic Coverage" with Allison (Pinosky) Gaines and Todd Murphey!

My ICRA Keynote: “Robot Learning in Isolation” This keynote talk is about technical paths for systems that do not have communications and are operating in novel environments, building machine learning models from scratch as they go. youtube.com/watch?v=hcYrL8…

I am super proud of this paper. All of you who have heard me rant about evaluation in robot learning papers - this is what I'd like to see from everyone! My favorite plots - Figures 4,6 (well, and 5). Violin plots of the *full data* (task completion), not just averages.