Research interests- Decision theory
- Reinforcement learning and Control
- Machine learning and Statistics
- Security, Privacy and Fairness
Open Projects- Open PhD position in multi-agent AI, mechanism design and differential privacy at Lille, France. [in collaboration with Chalmers and Harvard]. - A PhD position in reinforcement learning, under information and computation constraints at Chalmers, Sweden [Apply]- A Postdoc and a Research Engineer position for big data analysis in autonomous vehicles, at Chalmers, Sweden, in collaboration with Volvo cars are available immediately. - Contact me for possible collaborations or projects.Current group members and collaborators- Aristide Tossou (Chalmers), PhD student on Secure and Private Machine Learning and Decision Making.
- Hannes Eriksson (Chalmers), Research Engineer.
- Raphael Duroselle (Ecole Polytechnique), Intern.
- David Parkes (Harvard), Professor.
- Paul Tylkin (Harvard), PhD student (David Parkes).
- Goran Radanovic (Harvard), Postdoc.
- Yang Liu (Harvard), Postdoc
- Maryam Kamgarpour (ETHZ), Assistant Professor
- Nikolaos Tziortziotis (Ecole Polytechnique), Postdoc.
Teaching | News - At Harvard until 26 July 2017.
- Workshop in Theory and Practice of Differential Privacy, at ICML 2016.
- Open PhD position: Swedish Research Council Grant on "Learning, Privacy and the limits of Computation"
- Open PhD position: Future of Life Grant on Mechanim Design for Multiple AIs.
- Associate professor at INRIA-Lille and Univ-Lille from Oct 15 2015.
| Selected recent papers- Learning to Match, Value-aware and multi-stakeholder recommendation workshop at RecSys-17,
- Calibrated fairness In bandits, FATML-17.
- Subjective fairness: Fairness is in the eye of the beholder (working paper), 2017.
- Bayesian inference for Least Squares Temporal Difference regularization, ECML-17 (to appear).
- Differential privacy for Bayesian Inference through posterior sampling, JMLR, 2017.
- Achieving privacy in the adversarial multi-armed bandit, AAAI 2017.
- Thompson sampling for stochastic bandits with graph feedback, AAAI 2017.
- On the Differential Privacy of Bayesian Inference, AAAI 2016.
- Algorithms for Differentially Private Multi-Armed Bandits, AAAI 2016.
- Expected loss bounds for authentication in constrained channels. Journal of Computer Security, 2015.
- Cover tree Bayesian reinforcement learning. Journal of machine learning research, 2014.
- Personalized news recommendation with context trees, RecSys 2013.
- Probabilistic inverse reinforcement learning in unknown environments, UAI 2013.
- ABC reinforcement learning, ICML 2013.
|
