I am happy to share that our paper with @domienschepers and @vanhoefm got accepted to AsiaCCS 2019. The paper demonstrates new attacks against WPA-TKIP and is yet another evidence that there is no way to secure TKIP and we should all move on.
Sharing my winter break project. I tried combining @hardmaru's world models and @yaroslav_ganin's SPIRAL to see if an agent can learn to paint inside its own dream. It can! These strokes are generated purely inside a world model, yet transfer seamlessly to a real paint program.
Wanna contribute to the future of exoplanet research? Postdoc position open at @UniversiteLiege to develop focal-plane wavefront sensing for exoplanet imaging based on machine learning techniques, with application to @ELT_METIS: https://t.co/nkloRUb78D
Wonderful keynote talk by Johann Brehmer at #ACAT2019, strongly recommend checking his slides and papers https://t.co/n5XnbRUyhx
Several talks on mentioning LFI in this conference, so glad to see #HEPML focus increasingly shifting towards inference tasks from s/b classification
First paper of the year, by my student @joeri_hermans! In https://t.co/tVlhC96kRA, we adapt MCMC algorithms to the likelihood-free scenario by making use of a parameterized likelihood ratio model learned by supervised learning.
Our paper on how dendritic microcircuits approximates backprop was just accepted as an oral presentation for #NIPS2018 (only 30/4800 submissions were selected for an oral)! Fantastic job by Joao Sacramento (w/ Yoshua Bengio and Walter Senn).
On Thursday Aug 30 I'll be giving a talk at the @CERN Machine Learning forum about our recent work "Learning the likelihood function: moving from classification to inference" #HepML
https://t.co/yP9B1XB7at
https://t.co/pxtLfn59xn
https://t.co/iAqxK042ms
https://t.co/9ArguVxhov