Post-Doctoral Research Visit F - M Learning-Augmented Online Decision-Making H/F - INRIA

A propos d'Inria

Inria est l'institut national de recherche dédié aux sciences et technologies du numérique. Il emploie 2600 personnes. Ses 215 équipes-projets agiles, en général communes avec des partenaires académiques, impliquent plus de 3900 scientifiques pour relever les défis du numérique, souvent à l'interface d'autres disciplines. L'institut fait appel à de nombreux talents dans plus d'une quarantaine de métiers différents. 900 personnels d'appui à la recherche et à l'innovation contribuent à faire émerger et grandir des projets scientifiques ou entrepreneuriaux qui impactent le monde. Inria travaille avec de nombreuses entreprises et a accompagné la création de plus de 200 start-up. L'institut s'eorce ainsi de répondre aux enjeux de la transformation numérique de la science, de la société et de l'économie.
Post-Doctoral Research Visit F/M Learning-augmented online decision-making
Le descriptif de l'offre ci-dessous est en Anglais
Type de contrat : CDD

Niveau de diplôme exigé : Thèse ou équivalent

Fonction : Post-Doctorant

Niveau d'expérience souhaité : Jeune diplômé

A propos du centre ou de la direction fonctionnelle

The Inria Saclay-Île-de-France Research Centre was established in 2008. It has developed as part of the Saclay site in partnership with Paris-Saclay University and with the Institut Polytechnique de Paris .

The centre has , 27 of which operate jointly with Paris-Saclay University and the Institut Polytechnique de Paris; Its activities occupy over 600 people, scientists and research and innovation support staff, including 44 different nationalities.

Contexte et atouts du poste

Thepostdoc will be hired by Inria and integrate the FairPlay team, ajoint team with Criteo,ENSAE, and Inria. The team is hosted at CREST and visits Criteo weekly and the postdoc will be able to do the same and to interact with members of both institutions.

Mission confiée

Context

Online decision-making is a foundational scientific challenge applied in complex environments like dynamic routing and real-time advertising. Historically, designing algorithms for these systems has followed two divergent paths: theoretical worst-case competitive analysis, which provides strict guarantees but is often overly pessimistic, and purely data-driven machine learning methods (like reinforcement learning or bandit methods) that excel in average-case scenarios but can fail when facing out-of-distribution events or discrepancies between simulations and reality. To bridge the gap between these approaches, a powerful unifying paradigm has emerged: learning-augmented algorithms (or informed decision-making). This scientific framework equips online algorithms with machine-learned "advice" or predictions about the future, striving to create systems that are highly consistent when predictions are accurate, yet remain robust when the advice is limited, inexact, or adversarial.

Objectives

The primary objective of this project is to expand the theory of learning-augmented decision-making by systematically integrating realistic advice into fundamental online combinatorial optimization problems, such as scheduling and matching. The candidate will design algorithms that achieve optimal consistency-robustness trade-offs, identify which forms of advice are most valuable, and explore the strategic and game-theoretic implications of providing imperfect predictions to self-interested agents. Furthermore, the project aims to understand the computational costs of generating this advice from partial observations and capture the interaction between advice generation and its usage. Finally, these theoretical frameworks will be tested and validated by applying them to real-world industrial data, specifically aiming to design improved bidding policies in collaboration with Criteo.

Collaborations:

The post-doc will work with the permanent members of the FairPlay team, in particular Patrick Loiseau, SimonMauras and Vianney Perchet, but collaborations with other members of the FairPlay teamarewelcome.

Principales activités

Main activities: Understanding the state-of-the-art, developing novel algorithms and proving theoretical guarantees, developing and testing the solutions, communicating the work (papers, talks, etc.).

Compétences

Technical skills and level required: PhD level in the fields mentioned above fields

Languages: English mandatory, French is not mandatory

Relational skills: Taste for collaborative research

Other valued appreciated: Interest in theory and applications to online marketplaces

Avantages

- Subsidized meals
- Partial reimbursement of public transport costs
- Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
- Possibility of teleworking (after 6 months of employment) and flexible organization of working hours
- Professional equipment available (videoconferencing, loan of computer equipment, etc.)
- Social, cultural and sports events and activities
- Access to vocational training
- Social security coverage

Rémunération

Monthly gross salary : 2.788 euros