Maria Attarian

Research @GoogleDeepMind #GoogleBrain, PhD student @UofTRobotics @VectorInst ❤️ robots, IL and seeing the world through my camera lens. 🤖🌈🎠🎡🍭🍬🎈🎉🦄

[Maria Attarian @JMarakiii]

6/ Cool collaboration with Ian Vyse, @c_voelcker, @GerigkJasper, @evoprysh, @anasalmasr1, @Sumeet_Robotics, @du_yilun, @igilitschenski between @GoogleDeepMind, @UofT, @VectorInst, @UAlberta, @UTAustin and @Harvard.

[Maria Attarian @JMarakiii]

1/ Excited to be presenting our paper UF-OPS! 🎉🎉🎉

We present a method for test-time verifier optimization that can be applied to any stochastic policy and shows compelling improvements, in both sim and real!

Paper: https://t.co/VrqTT4EbD0

[Maria Attarian @JMarakiii]

5/ Real demo!

[Max Sobol Mark @maxsobolmark]

Joint work with @jackyliang42, @JMarakiii, @ChuyuanFu, @debidatta, @aviral_kumar2, and @shahdhruv_. Thanks to the GDM Robotics team!
Paper: https://t.co/EZta0uC92N
Website: https://t.co/5A2jfO6YAJ

[Google DeepMind @GoogleDeepMind]

We’re bringing powerful AI directly onto robots with Gemini Robotics On-Device. 🤖

It’s our first vision-language-action model to help make robots faster, highly efficient, and adaptable to new tasks and environments - without needing a constant internet connection. 🧵

[Vidhi Jain @viddivj]

I worked on Vid2Robot https://t.co/ufFHK1Dgbg as a student researcher at @GoogleDeepMind Robotics with @debidatta. This is a joint work with @JMarakiii, @nikhil_j_joshi, @ayzwah @DannyDriess @QuanVng @pannag_ Stefan Welker, Christine Chan, @igilitschenski, @ybisk.

[Pablo Samuel Castro @pcastr]

@davidusher This tool was based on a model @JMarakiii and I worked on:
https://t.co/vt5XNiODOi

And here's a video on the process, with @davidusher and @hugo_larochelle :
https://t.co/l16lxPZGSe

[Lily Goli @lily_goli]

Happy to announce that “Bayes’ Rays: Uncertainty Quantification for Neural Radiance Fields” is accepted to #CVPR2024 !

Thanks to my wonderful collaborators @code_red7777 @sellan_s @_AlecJacobson @taiyasaki.

[Jacky Liang @jackyliang42]

We can teach LLMs to write better robot code through natural language feedback. But can LLMs remember what they were taught and improve their teachability over time?

Introducing our latest work, Learning to Learn Faster from Human Feedback with Language Model Predictive Control

[University of Toronto @UofT]

#UofT researchers develop video camera that captures 'huge range of timescales' ➡️ https://t.co/YJ5UkV6D1u

[Maria Attarian @JMarakiii]

6/ Cool collaboration with @a_gi_dil, @JasonJZLiu, @RuthrashH, @animesh_garg, @igilitschenski, @JonathanTompson between @GoogleDeepMind, @UofT and @GeorgiaTech. Come to our poster this afternoon at 2:45pm to learn more!

[Maria Attarian @JMarakiii]

1/ Excited to be presenting our paper GeoMatch at #CoRL2023! 🎉🎉🎉

We train a unified grasping policy across multiple embodiments with various degrees of freedom that performs well across embodiments on unseen objects.

Paper: https://t.co/5XgCM8PkiZ

[Maria Attarian @JMarakiii]

5/ Real demo!

[Yuki @y_m_asano]

We've now released the @ICCVConference 's BigMAC workshop recording: https://t.co/oQJpwDzuUI

Also: most of the speakers' slides are now on the website too (and we're chasing the remaining ones) :).