I give a tentative theoretical justification for why large overparametrization is important in neural networks.

Primarily based on "A Universal Law of Robustness via Isoperimetry" by S.B. and Mark Sellke.
https://arxiv.org/abs/2105.12806

Project page for 'Omnimatte: Associating Objects and Their Effects in Video.'

Project website: https://www.computationalimaging.org/publications/keyhole-imaging/

Non-line-of-sight (NLOS) imaging and tracking is an emerging technology that allows the shape or position of objects around corners or behind diffusers to be recovered from…

It is not the fault of Turing, though.

cryptographic protocol for two-party computation

Short video of our new work on safe reinforcement learning in robotics. Accepted at CoRL for oral talk (6.5% acceptance rate). Share your comments of what you think!!

The full paper can be found at: https://openreview.net/forum?id=zwo1-MdMl1P

This video accompanies the paper

Chris Yu, Henrik Schumacher, Keenan Crane
"Repulsive Curves"
ACM Transactions on Graphics (2021)

For more information, see http://www.cs.cmu.edu/~kmcrane/Projects/RepulsiveCurves/

Shape from Metric project page

The Nobel Prize in Physics 2021 was awarded "for groundbreaking contributions to our understanding of complex physical systems" with one half jointly to Syukuro Manabe and Klaus Hasselmann "for the physical modelling of Earth’s climate, quantifying variability…