This was a very fun project to work on as it represents a new connection between the fields of optimization and circuit theory.
You can check out the paper here:
openreview.net/forum?id=9Jm...
(3/3)
Last week, I visited UCSD under Prof. Yang Zheng’s invitation to speak about this work. The following is a re-recording of the talk:
youtu.be/NcI-qxTqgn8?...
(2/3)
Your optimization algorithm is a circuit!
In our NeurIPS 2024 Spotlight paper with Stephen Boyd, Tetiana Parshakova, Jaewook Suh, we present a framework for designing optimization algorithms using (electric) circuit theory.
(Video and paper links in reply)
Yesterday, Kevin Shu (Caltech) visited UCLA and gave a wonderful talk at the Mathematics Colloquium titled Algebraic Methods in Convex Optimization!
You can check out the recording here:
youtu.be/e10y8fJdZ2Y
In that sense, this is control theory. The authors have EE backgrounds, so we were more comfortable expressing things with RLC components rather than, say, springs and dampers.
Optimization Algorithm Design via Electric Circuits @StephenPBoyd1
Aaron gave a truly amazing talk. Very clear and super informative 👍
youtu.be/QUgi3-XuZ3g
Shuvomoy Das Gupta (Columbia) and I (UCLA) are starting an optimization seminar series! Our first speaker, Aaron Defazio (Meta), presented Schedules & Schedule-Free Learning.
Aaron will give his NeurIPS 2024 Oral work next week. This is a longer version. (Video link in reply)
Can you add me plz? Thanks!