Papers of the day   All papers

Limitations of adversarial robustness: strong No Free Lunch Theorem

Comments

Apr 23 2019 Elvis Dohmatob

Caution: There will be likely No Free Lunch in #ICML2019 ! My theoretical paper on the impossibility of adversarial robustness accepted :) Please RT :) Slides: https://drive.google.com/file/d/1IOmZqLtujaqIywdPMtEOde6pT2bnlbg4/view?usp=sharing Preprint: https://arxiv.org/pdf/1810.04065.pdf https://t.co/2L5gasNZci
7 replies, 385 likes


Jun 19 2019 Elvis Dohmatob

1/ Apparently, I'm ranked 6th in the obscure tail of single-authors @ #ICML2019 / @icmlconf. Ashok Cutkosky from Google is 1st. Welcome to the Darkside! See reddit (r2drkqkkeix21.png) for full post. Slides: https://icml.cc/media/Slides/icml/2019/grandball(11-11-00)-11-12-10-4738-generalized_no_.pdf Paper: https://arxiv.org/pdf/1810.04065.pdf https://t.co/LUpDqkyYMd
4 replies, 35 likes


May 23 2019 Elvis Dohmatob

In case you're in Paris on the 28th, I'll be giving an hors-serie presentation of some recent work (on adversarial examples) at the Paris ML meetup. Register here: https://www.meetup.com/Paris-Machine-learning-applications-group/events/261543471/ Ref: https://twitter.com/dohmatobelvis/status/1120621890601725952 @bardolfranck @IgorCarron
2 replies, 27 likes


May 24 2019 Criteo AI Lab

With 5 papers accepted at @icmlconf we are looking forward to meeting, sharing and breaking new ground with the community in Long Beach! #icml2019
0 replies, 10 likes


May 09 2019 Daniel Bilar

@cigitalgem @SaumyaJetley See re ML adversarial robustness: Strong No Free Lunch Theorem by @dohmatobelvis "sharp & definitive limit to the amount of robustness that can be guaranteed" https://twitter.com/dohmatobelvis/status/1120621890601725952 A #ICLR2019 paper IIRC
1 replies, 4 likes


May 29 2019 Elvis Dohmatob

2/ My fundamental work https://arxiv.org/pdf/1810.04065.pdf (yes it, generalizes everything known on the subject) on the general impossibility of adversarial robustness makes heavy-duty use of his ideas, and extensions thereof developed by @VillaniCedric in the 2000's (HWI inequalities).
0 replies, 1 likes


Jun 01 2019 Arthur Charpentier 🌻

"Limitations of adversarial robustness: strong No Free Lunch Theorem" https://arxiv.org/pdf/1810.04065.pdf
0 replies, 1 likes


Content