Konstantinos Pitas, Ph.D.

Inria Grenoble Rhône-Alpes, Statify Team.

prof_pic.jpg

655 Av. de l'Europe

Montbonnot-Saint-Martin

38330, France

I’m a postdoctoral researcher at Inria Grenoble Rhône-Alpes, Statify team. In 2020, I completed my Ph.D. at École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, which I obtained under the supervision of professor Pierre Vandergheynst. I completed my Bachelor’s and Master’s in Electrical and Computer Engineering at Aristotle University of Thessaloniki (AUTh), Greece.

My research interests include broadly various topics in ML Safety, with a particular focus on Deep Ensembles, Bayesian Neural Networks and PAC-Bayesian generalization bounds. I am also interested in LLMs and GNNs.

I’m very passionate about transparent governance and enjoy analysing various topics on political polling and psephology on kassiope.org (in Greek).

News

Oct 4, 2023 Something for (almost) nothing: Improving deep ensemble calibration using unlabeled data, new Arxiv preprint
Sep 11, 2023 The fine print on tempered posteriors, accepted at ACML2023! :sparkles: :smile:
Jun 1, 2023 Europepolls: A Dataset of Country-Level Opinion Polling Data for the European Union and the UK, new Arxiv preprint
Jun 22, 2022 Cold Posteriors through PAC-Bayes, accepted at NeurIPS 2022 Workshop on Trustworthy and Socially Responsible Machine Learning

Activities