Home

[Open positions at CEA in Formal Methods and AI]

For a bit of history, I joined CEA in 2015, at the end of my PhD, for a postdoc on the Soprano project. (No, no, it is unrelated to this soprano... nope, not really to this soprano either... good Gödel definitely NOT this soprano... Oh I wish it was this soprano... Okay I'll stop now, lest I become an unfortunate character in this soprano...)

I was a PhD student working on the ProofCert project in the Parsifal team (Inria Saclay) registered at EDX (École polytechnique) under the supervision of Dale Miller. I graduated November 2015, and I will not fix your machine.

I am generally interested in logic and AI, and particularly in certification, theorem proving and proof checking, software verification & validation, machine learning, higher order logic programming, focusing. You could even say I am married to Focused Intuitionistic Linear Higher-Order Logic. And I am always curious and looking for new research adventures! I touched on many areas during my studies including AI, algorithmics, functional programming, automata, semantics, and each gave me great pleasure.

I am also passionate about the History of Logic, how the fields and discoveries influenced each other, how the collision of paradoxes with foundations spawned new fields, and how these, in turn, permeated and, sometimes gave birth to, most of computer science. The intricacies of the endless creation of knowledge. The phiLOGenetICs of Logic, if you will. But, to be honest, I am also interested in the gossip, so please do share with me any cool anecdote you might hear, like the time Poincaré poked Russell in the ribs and Russell answered "This is foolish, I say this quite deliberately", or the time Leibniz said to Newton "Finders keepers, losers weepers"... No? Am I getting these wrong again...?

And finally (for a little fun!), I recently published, under the pen-name Hizachi, the French adaptation of Cryptic Crosswords, named Crucicryptos, which you can know all about by visiting www.crucicryptos.fr. You can also get regular previews by joining the Facebook page fb.com/crucicryptos and by following twitter.com/crucicryptos. Or... more simply... you can just get the book at Amazon.fr, Amazon.de, Amazon.ca, Amazon.co.uk or Amazon.com.

I am currently a researcher at the Safety and Security Lab of CEA-list, where I started, and now lead, the effort toward the safety ensurance of Artificial Intelligence (not limited to Machine Learning), particularly through Formal Methods (although we are always open for new techniques). The techniques we currently use include Abstract Interpretation, SMT solving and Constraint Programming, and we are developping new tools as well as adapting existing ones.

My full name is Zakaria Hichem Chihani but if that's unpronounceable in your mother tongue, feel free to call me Zak. My family name is officially pronounced "/šiḥani/", with a guttural "h". But feel free to say "chiHani" with an "h" like in "haha", or "shianee" like the French people expediently pronounce it, or "kiaaaaaaany" in an Italian accent. Or... you know... Zak.