About Me

I am a postdoc researcher at EPFL (Switzerland), collaborating with Carmela Troncoso at the SPRING lab. I did a PhD in Machine Learning and Information Security at Royal Holloway University of London with the CDT programme, supervised by Alex Gammerman, advised by Kenny Paterson. I received an MSc in Machine Learning from Royal Holloway University of London in 2014, and a BSc in Mechatronics and Computer Engineering from University of Pavia in 2013.


Email: <name>.<surname>@epfl.ch

GitHub: gchers. Twitter: @gchers.

GPG: 0xC31273819075CF37

Research Interests

I generally try to work on projects involving a good mixture of theory and practice.

My current research aims at measuring systems’ leakage by using methods from the Machine Learning theory; I applied this to side channel attacks (e.g., traffic analysis). I also worked on extending methods for confident prediction (Conformal Predictors), by proposing: a clustering algorithm, an alternative to the List-Viterbi algorithm for Hidden Markov Models, and methods to form ensembles of Conformal Predictors.

My main interests are in the following areas:

I play with the CTF team of RHUL, TU6PM.

I am a (happy) OpenBSD and FreeBSD user, and I would highly recommend you become too.

I sporadically write words on topics I find interesting.

Publications

Majority vote ensembles of conformal predictors. Giovanni Cherubin. Machine Learning, Aug 2018. [ DOI | http ]

Exchangeability martingales for selecting features in anomaly detection. Giovanni Cherubin, Adrian Baldwin, and Jonathan Griffin. In Proceedings of the Seventh Workshop on Conformal and Probabilistic Prediction and Applications, volume 91 of Proceedings of Machine Learning Research, pages 157--170. PMLR, 2018. [ code | pdf | .html ]

Bayes, not Naïve: Security Bounds on Website Fingerprinting Defenses. Giovanni Cherubin. Proceedings on Privacy Enhancing Technologies, 4:215--231, 2017. [ video | code | slides | pdf ]

Website Fingerprinting Defenses at the Application Layer. Giovanni Cherubin, Jamie Hayes, and Marc Juarez. Proceedings on Privacy Enhancing Technologies, 2:186--203, 2017. [ code | pdf ]

Hidden Markov Models with Confidence. Giovanni Cherubin and Ilia Nouretdinov. In Conformal and Probabilistic Prediction with Applications - 5th International Symposium, COPA 2016, Madrid, Spain, April 20-22, 2016, Proceedings, pages 128--144, 2016. [ DOI | slides ]

Conformal Clustering and Its Application to Botnet Traffic. Giovanni Cherubin, Ilia Nouretdinov, Alexander Gammerman, Roberto Jordaney, Zhi Wang, Davide Papini, and Lorenzo Cavallaro. In Statistical Learning and Data Sciences - Third International Symposium, SLDS 2015, Egham, UK, April 20-23, 2015, Proceedings, pages 313--322, 2015. [ DOI | slides ]

Other

F-BLEAU: Practical Channel Leakage Estimation. Giovanni Cherubin, Kostas Chatzikokolakis, and Catuscia Palamidessi. Under sumbission, 2018. [ code ]

Bots detection by Conformal Clustering. Giovanni Cherubin. Master's thesis, Royal Holloway University of London, 2014. [ pdf ]

Slides

Measuring the Leakage of a Black-box using Machine Learning. 2018. Alan Turing Institute, London. [ video | .pdf ]

Exchangeability martingales for selecting features in anomaly detection. 2018. Workshop on Conformal and Probabilistic Prediction and Applications. [ .pdf ]

On the Security Against Machine Learning-based Attacks. 2017. CDT Showcase, Evelyn Sharp Centre, Sunningdale Park. [ .pdf ]

Bayes, not Naïve: Provable Security of Website Fingerprinting Defenses. 2017. ISG Seminar, Royal Holloway University of London, UK. [ .pdf ]

Applications of Conformal Prediction in Information Security Problems. 2016. CDT Showcase, Windsor Great Park, UK. [ .pdf ]

Conformal Clustering and Bots Traffic. 2015. (Invited) CPRML Workshop 2015, Hyderabad, India. [ .pdf ]

Research Visits

Research Engineer, HP Labs Security Lab, Bristol (August-November 2017)
Supervisors: Jonathan Griffin, Adrian Baldwin

Research Visitor, École Polytechnique, Paris (May; November 2017)
Supervisors: Prof. Catuscia Palamidessi, Kostas Chatzikokolakis

Research Intern, Cornell Tech (June-September 2016)
Supervisor: Prof. Thomas Ristenpart

Academic Service

Teaching

I have been teaching assistant for R programming for the courses on Machine Learning and Data Analysis at Royal Holloway University of London (2014-17). I was teaching assistant for the courses on C programming and Linear Algebra and Geometry at University of Pavia (2011-12).

Reviewing

I am a PC member for COPA 2018. I am external reviewer for PETS (2017-18), and have been external reviewer for Financial Cryptography (2017).

Awards

Year Award Category
2017 Best Paper Andreas Pfitzmann Best Student Paper Award at PETS: “Bayes, not Naïve: Security Bounds on Website Fingerprinting Defenses”
2017 Prize First place at Capture The Flag (CTF) security challenge organised by NCC Group at the Cambridge2Cambridge event
2015 Best Paper Best student paper award sponsored by HP at SLDS conference: “Conformal Clustering and Its Application to Botnet Traffic”
2014 Best Finalist Best MSc Big Data finalist in memory of Prof. Alexey Chervonenkis (Royal Holloway University of London)