---

Here are some research projects I advised recently at the graduate level at JIE:

• 2016-2017: Deep reinforcement learning
• 2016-2017: Adaptive video streaming via reinforcement learning
• 2016-2017: Reinforcement learning for UAV control
• 2016-2017: Automatic trading with reinforcement learning
• 2016-2017: Adaptive navigation in transportation networks
• 2015-2016: Finding risk-averse shortest path using time-dependent stochastic costs
• 2015-2016: A multi-objective reinforcement learning approach to traffic light control in nonstationary environments
• 2015-2016: Discrete and continuous decoding of reaching movements from neural signals
• 2015-2016: Optimized spaced repetition based on a learner's model
• 2015-2016: Batch reinforcement learning for optimized spaced repetition
• 2015-2016: Empirical evaluation of Markov decision process and reinforcement learning with ordinal rewards

Here are some research projects I advised at the graduate level at JRI:

• 2016-2017: Combining shortest path and recommender system
• 2016-2017: Algorithmic decision theory in computer networks

Here are some research projects I advised at the graduate level (master MVA) at ENS Cachan:

• 2016-2017: Inverse reinforcement learning with additional preferential information
• 2015-2016: Community detection based on mixing times

Here are some research internships I advised at the graduate level (M2) at UPMC:

• 2013: Solving optimization problems in poor information situations (with Thibaut Lust)
• 2013: Sequential decision-making with ordinal rewards (with Paolo Viappiani)
• 2013: Parameterizing preference models for sequential decision-making under uncertainty (with Paolo Viappiani)
• 2012: Policy learning in unknown environment by regret minimization (with Aurélie Beynier)
• 2011: Reinforcement learning for dynamic asset allocation (at Primexis)
• 2009: Analysis of Schelling's model (mainly advised by Cyril Banderier)

Here are some research projects I proposed at the graduate level (M2) at UPMC:

• 2014: Combinatorial adversarial multi-armed bandits
• 2014: Preference-based policy learning
• 2013: Multiobjective reinforcement learning
• 2013: CMA-ES (Covariance Matrix Adaptation - Evolution Strategy)
• 2007: Decision-making with partially consonant belief functions

Here are some programming projects I proposed at the graduate level (M1):

• 2013: Trading on the bitcoin market
• 2013: Reinforcement learning agent with a tutor (with Paolo Viappiani)
• 2012: Application in finance of robust or risk-sensitive reinforcement learning
• 2011: Backtests of financial investment strategies
• 2011: Artificial Intelligence and Poker Texas Hold'em
• 2010: Pair-trading strategies based on copulas
• 2009: Study of solution methods for large-scale MDPs (with Aurélie Beynier)