Research

Working Papers: 

• Achieving O (1/N) Optimality Gap in Restless Bandits through Diffusion Approximation: with Weina Wang and Lei Ying   [arXiv] 

• Certainty Equivalence Control-Based Heuristics in Multi-Stage Convex Stochastic Optimization Problems: with Alexandre REIFFERS-MASSON   [arXiv] 

• The LP-Update Policy for Weakly Coupled MDPs: with Nicolas GAST and Bruno GAUJAL   [arXiv] 

• ANalysis Of Variability of EXtremes: Combing ANOVEX with Decision Trees: with Thomas OPITZ, Stéphane GIRARD and Antoine USSEGLIO-CARLEVE
  This manuscript summarizes the research from my current post-doc position. While it is still in draft form, slides that outline the prerequisites and main ideas can be found here .  

Research Papers: 

• An Optimal-Control Approach to Infinite-Horizon Restless Bandits: Achieving Asymptotic Optimality with Minimal Assumptions [arXiv] (accepted and will be presented in the 2024 CDC conference)

• LP-Based Policies for Restless Bandits: Necessary and Sufficient Conditions for (Exponentially Fast) Asymptotic Optimality: with Nicolas GAST and Bruno GAUJAL   [arXiv] (published in Mathematics of Operations Research)

• Exponential Asymptotic Optimality of Whittle Index Policy: with Nicolas GAST and Bruno GAUJAL   [published in Queueing Systems volume 104, pages107–150 (2023) ]  [arXiv] 

Talks and Presentations:

0. The presentation of my PhD defence can be found here ; A video of the defence can be found here 

1. Diffusion-Based Policy for Restless Bandits: INFORMS Annual Meeting 2024, Seattle, USA   [Slides] [Poster]

2. Asymptotic Optimality in the Markov Bandits: Assumptions and the Rate of Convergence: Workshop on Reinforcement Learning for Stochastic Networks, Toulouse, France  [Slides]

3. Certainty Equivalence Control in Restless Bandits: Workshop on Restless Bandits, Index Policies and Applications in Reinforcement Learning, Grenoble, France   [Slides]

4. ANOVEX Applied to Change Point Estimation in the Extremes: exCRG online workshop between KAUST, Lancaster Universities and INRAE   [Slides][Video]

5. ANalysis Of Variability of EXtremes: Combing ANOVEX with Decision Trees: Extreme Value Analysis (EVA) 2023, Milan, Italy   [Slides] 

6. The LP-update Policy for Weakly Coupled Markov: A Near-Optimal Re-Solving Heuristic: La ROADEF (Société Française de Recherche Opérationnelle et d'Aide à la Décision) 2023, Renne, France   [Slides] [Text]

7. LP-based Policies for Restless Bandits: Conditions for (Exponentially Fast) Asymptotic Optimality: POLARIS team presentation   [Slides] [Video]

8. Asymptotic Optimality of Index Policies for Markovian Bandits: La ROADEF (Société Française de Recherche Opérationnelle et d'Aide à la Décision) 2021, Online  [Slides] [Text]

9. Asymptotic Optimality of Index Policies for Markovian Bandits: Journée de Laboratoire d'Informatique de Grenoble, France, 2021   [Poster]

10. Restless Bandits and Mean Field: Master 2 internship presentation, 2020   [Slides] [Text]