sorina
@robot_in_space2
Research Fellow @ Openmind Research Institute || PhD @ Caltech || Keck Institute of Space Sciences (KISS) Affiliate || NASA-JPL Affiliate || EPFL alumna || 🇹🇩
Happy New Year from our robots to yours! 😻 youtu.be/wBJqv6rsT_k?si…
His lectures and explanations are amazing! I learned a lot of controls from his Control Bootcamp!
New Book & Video Series!!! (late 2025) Optimization Bootcamp: Applications in Machine Learning, Control, and Inverse Problems Comment for a sneak peak to help proofread and I'll DM (proof reading, typos, HW problems, all get acknowledgment in book!)
I love papers that are rigorous and thoughtfully written. I haven’t read this one yet, but I know it’s going to be excellent. Highly recommend giving it a look!
TRI's latest Large Behavior Model (LBM) paper landed on arxiv last night! Check out our project website: toyotaresearchinstitute.github.io/lbm1/ One of our main goals for this paper was to put out a very careful and thorough study on the topic to help people understand the state of the…
A lot of people, especially practitioners, think they understand RL because they know PPO (or DDPG or SAC). I used to be in the same boat a couple of years back. But that’s a narrow slice of the field. There's so much more to RL, and it's extremely rewarding. Happy 4th of July!…
Coding with AI can feel great, but LLMs are unreliable and introduce subtle, hard-to-detect bugs. When you're tired, it's especially easy to accept or produce flawed code without realizing it. In addition, the human brain tends to see what it expects rather than what’s actually…
Our poster received the Runner-Up Award (2nd place) for Best Poster at the #RSS2025 Workshop on Resilient Off-road Autonomous Robotics! What makes me happier: my husband’s poster won Best Poster (1st place). Huge congratulations to Patrick @Stapelzeiger 💙 and his RACER team.…
I really like this diagram, it's from a presentation by Prof. Jonathan Lunine @jlunine at the Keck Institute for Space Studies + my annotations in black.
We won the Best Paper award at the #RSS2025 Space Robotics Workshop!
Yes! Along the same lines, read control theory textbooks. Personally, I find many recent papers somewhat convoluted (especially in ML), whereas older books and papers tend to be more clearly structured and use less unnecessary fancy words.
All the more so.
Falcon 9 rocket launch as seen from @Caltech’s construction site. Space exploration is magical 🚀
I was so fortunate to work with such an amazing advisor at @Caltech, Prof. Soon-Jo Chung. Thank you for all the support and I apologize for almost burning your lab down twice through my experiments. ❤️
I am so proud of my undergrads, Sulekha and Leo, for being featured in the @Caltech magazine. ❤️
Ten of Caltech’s 2025 graduates discuss their achievements, challenges, and lessons learned during their time at the institute. #Caltech2025 magazine.caltech.edu/post/socaltech…
Thesis deposited! I poured my heart into this work, and now that it’s done, I miss it. This thesis was more than just a research project. It was a chance to chase ideas I believed in to change the world: – vision-based control and real-time learning – sending learning…