
Mahalakshmi Sabanayagam
@maha81193
PhD student @ TU Munich. Interested in theory of deep learning and graph based learning.
ID: 742140770
https://mahalakshmi-sabanayagam.github.io/ 07-08-2012 04:28:59
298 Tweet
150 Takipçi
453 Takip Edilen


Receiver Operating Characteristic (ROC) got its name in WWII from Radar, invented to detect enemy aircraft and ships. ROC curve plots true pos. rate vs false pos. rate, parametrized by a detection threshold ROC has many interesting properties 1/8 animation by Dariya Sydykova

At TFAI TUM reading group, we discuss the paper "EARLY STOPPING IN DEEP NETWORKS: DOUBLE DESCENT AND HOW TO ELIMINATE IT" with one of the authors Reinhard Heckel on Wednesday, 24.11 @ 9:30CEST openreview.net/pdf?id=tlV90jv… Feel free to join us: tum-conf.zoom.us/j/66633677821 Passcode 537320





Our paper won the best paper award AdvMLFrontiers #NeurIPS2024 🎉 Special thanks to my collaborator Lukas Gosch!🙂 We derive the first white-box robustness certificates for (graph) neural nets against data poisoning & backdoor attacks. Check out the talk on 14 Dec NeurIPS Conference

Super happy & honored that our work on certifying NNs against poisoning won the Best Paper Award AdvMLFrontiers at #NeurIPS2024. Come by our poster 10:40am-12&4-5pm (or talk) tomorrow :) Joint work w/Mahalakshmi Sabanayagam, Debarghya Ghoshdastidar & Stephan Günnemann L: arxiv.org/pdf/2407.10867
