Hello, since Nov. 2018, I am on an “advanced research position” at Inria, in the WIDE team.
“Maybe, not content to be alienated by observation, maybe does the object deceive us? Maybe it invented original answers, and not only the ones we ask for? May not be does it not want to be analyzed and observed at all and taking this for a challenge (which is true), does it respond with another challenge? […] the object analyzed today triumphs everywhere, by its object position, over the subject of analysis.” (Jean Baudrillard, Fatal Strategies.)
My main current research activity is trying to disprove that quote, by studying black-box algorithms in the context of recommender systems, neural-network models, or decision-making algorithms in general. I co-organized with Gilles and Antoine the workshop “Algorithmes en Boite-Noire”, in a continuous attempt to link several research domains on this societal question.
- Article “Le problème du videur : la crédibilité des explications de l’IA en question” on interstices.info.
- I am the PI of the Audita associate team, btw EPFL (SaCS team) and Inria (WIDE), starting in 2021.
- I am at the scientific board of REGALIA.
- Paper accepted in CVPR 2021: SurFree: a fast surrogate-free black-box attack, presenting a fast (in number of queries) attack to find adversial examples.
- Paper accepted in INFOCOM 2021: Setting the Record Straighter on Shadow Banning, about the shadow banning practices in Twitter.
- Launched the whosban.eu.org website, to check how shadow banned you are on Twitter.
- Related to our black-box observations of YouTube recommendations
- New article in Nature Machine Intelligence, do not trust remote AI explanations… [Article link]
- pres: “Le problème du videur : la crédibilité des explications de l’IA en question” sur interstices.info.
- press: Do explanations for data-based predictions actually increase users’ trust in AI? Tech Xplore. Rebranded in Techhq.
- press: De l’explicabilité des systèmes : les enjeux de l’explication des décisions automatisées internetactu.net.
- press: Explainable AI” doesn’t work for online services – now there’s proof Algorithm Watch.
- press: It’s Too Easy to Hide Bias in Deep-Learning Systems IEEE spectrum.
- Juin 2020: participation à la task force “Régulation des GAFA / Transparence et audit des algorithmes” auprès de la Direction Générale des Entreprises (DGE).
- Semestre Supervision de systèmes dynamiques.
- Feb 2020: joined the board of the Société Informatique de France.
- Setting the Record Straighter on Shadow Banning, Erwan Le Merrer, Benoît Morgan, Gilles Trédan. In INFOCOM (2021). [IEEE link] [ArXiv preprint 2020].
- press: Exploring the underpinnings of shadowbanning on Twitter TechXplore.
The Imitation Game: Algorithm Selection by Exploiting Black-Box Recommenders, Georgios Damaskinos, Rachid Guerraoui, Erwan Le Merrer and Christoph Neumann. In NETYS, 2020. [Preprint].
zoNNscan : a boundary-entropy index for zone inspection of neural models, Adel Jaouen, Erwan Le Merrer. In Monte Carlo Search workshop (MCS) @ IJCAI 2020 [ArXiv preprint 2018].
- Uncovering Influence Cookbooks : Reverse Engineering the Topological Impact in Peer Ranking Services, Erwan Le Merrer, Gilles Trédan. In ACM CSCW, 2017. [preprint] [editor]
Remote Explainability faces the bouncer problem, Erwan Le Merrer, Gilles Trédan. Nature Machine Intelligence, 2020. [citation]
Adversarial Frontier Stitching for Remote Neural Network Watermarking, Erwan Le Merrer, Patrick Perez, Gilles Trédan. Neural Computing and Applications, 2020.
Frugal topology construction for stream aggregation in the cloud, Rachid Guerraoui, Erwan Le Merrer, Rhicheek Patra, Bao-Duy Tran. Infocom, 2016.
Archiving cold data in warehouses with clustered network coding, Fabien André, Anne-Marie Kermarrec, Erwan Le Merrer, Nicolas Le Scouarnec, Gilles Straub, Alexandre van Kempen. Eurosys, 2014.
Second order centrality: Distributed assessment of nodes criticity in complex networks, Anne-Marie Kermarrec, Erwan Le Merrer, Bruno Sericola, Gilles Trédan. Computer Communications, 2011.
- Thibault’s code for SurFree, a surrogate free black-box adversarial attack, [here].
- Arsany’s code for FeGAN and MD-GAN available here.
- Pushed non-randomness computation for graphs, by Ying & Wu, in NetworkX 2.4.
- zoNNscan code, for analysis of decision boundaries, available here.
- Second order centrality code now part of the Networkx library, from ComCom paper.
- Corentin’s code for distributed deep learning (NCA best paper) here and for random neural layers here.
I was a senior research scientist at Technicolor R&I (2009-2018), where I worked on scalable storage, processing and machine learning for data analytics and monitoring of home devices. I own a PhD on distributed systems from University of Rennes 1, and a background on peer-to-peer systems, and graph mining/algorithms. My Ph.D. thesis was financed by Orange Labs (where I stayed btw 2004-2007). I obtained my habilitation (HDR) on November 2016 from University of Rennes 1. I am the president and sysadmin of the gozdata association, that provides the gozmail service, built from free software. I am a board member of the Société Informatique de France.
- IEEE NCA 2017 best paper award, for paper “Distributed deep learning on edge-devices: feasibility via adaptive compression”, with Corentin Hardy and Bruno Sericola.
- Paper co-authors are awarded prize “La recherche 2015”, for work on cold storage, from Eurosys 2014.
- Prix “Bretagne jeune chercheur” 2011, mention spéciale.
- Talk at Forum Numerica jan’21: The Bouncer and the Boundary: Modern Decision-Making Algorithms and Their Explanations.
- Talk au THCON20 @ Toulouse avec G. Trédan.
- Atelier “algorithmes en boite-noire”, le 10 octobre 2019 à Lyon.
- Article Emergences/Inria “Plus de transparence pour les algorithmes de recommandation” par J-M Prima.
- Talk video about “Decision boundaries & security related questions” @ the Deep Learning : from theory to applications workshop in Sept. 2018 in Rennes: VIDEO HERE.
As my main participation to scientific events, I was co-chair of Middleware 2018 industry track, in Rennes, and AlgoTel 2019 program co-chair. Co-organizing WOS with INRIA and Interdigital (ex Technicolor), a workshop on storage, cloud processing and data analytics, in every year Rennes: WOS archive.
I have co-authored 8 granted patents, and around 39 pending demands.
I have contributed to produce a document on “Transparency and audit of algorithms” regarding the GAFAM, for the French DGE. In general, I am broadly interested in the impact of technology on society.