This tweet, from three years ago today, may be the most impactful tweet I've made. It led to this privacy mechanism w/ @ccanonne_ @shortstein https://
arxiv.org/abs/2004.00010 That mechanism was used in the 2020 US Census. If you were in the US on 4/1/2020, it was run on your data.
@thegautamkamath
-
Privacy Mechanism for 2020 US Census Data Protection
By
–
-
Reviewer Selection and Research Community Responsibility Practices
By
–
Combination of picking reviewers I know personally and/or with relevant expertise. I don't think I have any special formula that others wouldn't think of. I would give more credit to the pocket of the research community that I reside in, where people seem to be pretty responsible
-
Tim Roughgarden’s Inspiring Talk Captivates Full Audience
By
–
Such an inspiring and captivating talk by Tim Roughgarden. Full room, standing room only. I didn't see a single person on their phone or computer.
-
Tim Roughgarden Distinguished Lecture on Blockchains Web3
By
–
SUPER excited to have @Tim_Roughgarden visiting @UWCheritonCS today as our Distinguished Lecture Speaker! "The Long Arm of Theoretical Computer Science: A Case Study in Blockchains/Web3" Today at 3 PM! https://
cs.uwaterloo.ca/events/dls-tim
-roughgarden
… -
Membership Inference Attacks Applied to Beluga Whale Conservation
By
–
One of the most unique "privacy" papers I've ever seen: using membership inference attacks to discriminate between pictures of Beluga whales. Using privacy attacks for social good! Membership Inference Attack for Beluga Whales Discrimination https://
arxiv.org/abs/2302.14769 -
Reversible Vision Transformers: Memory-Efficient Deep Learning Architecture
By
–
I learned about this trick from this paper "Reversible Vision Transformers" by @Karttikeya_m et al (
https://
arxiv.org/abs/2302.04869). Not an expert here, but I think the reversible transformation is derived from previous work on NICE (
https://
arxiv.org/abs/1410.8516). 6/6 -
Memory Efficient Neural Networks: Activation Storage Optimization
By
–
This is useful for memory efficient NNs because with careful architecture choice, you don't have to store the activations of every layer. You can store the last layer and work your way backwards. The example indicates that the operation composes, so you can make deep networks 5/n
-
Mathematical Puzzle for Memory-Efficient Neural Networks
By
–
Cute puzzle, courtesy @Karttikeya_m
. Has applications to memory-efficient neural network architectures. Suppose F and G are arbitrary known functions (not necessarily invertible). You are given output O. Can you reconstruct input I? Hint in next tweet. 1/n -
Google Scholar Blog Contributors and AI Research
By
–
I thought that too, but maybe it's like, up to 5. Anurag Acharya is the name I associate with it. But there are a few more people who have written the blog posts here: https://
scholar.googleblog.com -
Google Scholar design control and alternative academic search platforms
By
–
Agreed. The design choices, which are norm setting, are at the whims of a couple of folks (
https://
x.com/thegautamkamat
h/status/1541059471249752066
…). There are alternatives (e.g., @SemanticScholar
). If GS went under, then I assume something else would take its place.