Phd Position F/m doct2024-neo Federated - Sophia Antipolis, France - Inria

Inria

Entreprise vérifiée

Sophia Antipolis, France

il y a 2 semaines

Posté par:

Sophie Dupont

beBee Recruiter

Description

Le descriptif de l'offre ci-dessous est en Anglais_

Type de contrat :

CDD

Niveau de diplôme exigé :
Bac + 5 ou équivalent

Fonction :
Doctorant

Niveau d'expérience souhaité :
Jeune diplômé

A propos du centre ou de la direction fonctionnelle:
The Inria centre at Université Côte d'Azur includes 37 research teams and 8 support services. The centre's staff (about 500 people) is made up of scientists of different nationalities, engineers, technicians and administrative staff.

The teams are mainly located on the university campuses of Sophia Antipolis and Nice as well as Montpellier, in close collaboration with research and higher education laboratories and establishments (Université Côte d'Azur, CNRS, INRAE, INSERM...), but also with the regiona economic players.

With a presence in the fields of computational neuroscience and biology, data science and modeling, software engineering and certification, as well as collaborative robotics, the Inria Centre at Université Côte d'Azur is a major player in terms of scientific excellence through its results and collaborations at both European and international levels.

Mission confiée:

The integration of federated learning (Kairouz et al., 2021) with reinforcement learning (Sutton and Barto, 2018) presents a promising avenue for developing decentralized, privacy-preserving machine learning models that are capable of real-time decision-making across distributed networks.

However, this integration faces challenges including, but not limited to, efficiency in training, client selection, convergence guarantee, data heterogeneity, and privacy concerns (see e.g.,Qi et al., 2021).

Much less research has been devoted on addressing potential statistical heterogeneity across agents, e.g., in terms of the evolution of their environments or of the rewards (Jin et al, 2022), (Xiong et al, 2024), (Zhang et al, 2024).

This PhD project aims to develop novel reinforcement learning algorithms tailored for correlated Markov process-driven agent environments, with a focus on their theoretical underpinnings.

The relevant research of the team on the reinforcement learning for restless multi-armed bandits and on personalized federated learning is (Avrachenkov and Borkar, 2022), (Marfoq et al., 2021), and (Marfoq et al., 2022).

References:

Richard S. Sutton and Andrew G.

Barto, Reinforcement Learning:
An Introduction, MIT press, 201- Peter Kairouz et al., Advances and Open Problems in Federated Learning, Now Publishers, 202- Jiaju Qi, Qihao Zhou, Lei Lei, Kan Zheng,

Federated Reinforcement Learning:
Techniques, Applications, and Open Challenges, Intelligent Robotics 2021;1(1):18-5- S.

Zhang, Jieyu Lin, Qi Zhang, A Multi-agent Reinforcement Learning Approach for Efficient Client Selection in Federated Learning, AAAI 202- Zhenan Fan, Yue Ning, Huzefa Rangwala, A Fair Federated Learning Framework With Reinforcement Learning, ArXiv 202- S.

Liu, Shiyuan Yang, Hanze Zhang, Weiguo Wu, A Federated Learning and Deep Reinforcement Learning-Based Method with Two Types of Agents for Computation Offload, MDPI Sensors 202- J.

Wang, J. Hu, J. Mills, G. Min, M. Xia, and N. Georgalas, Federated Ensemble Model-based Reinforcement Learning in Edge Computing, IEEE Trans on Parallel and Distrib uted Systems, Vol. 34, N. 6, June 202- H. Jin, Yang Peng, Wenhao Yang, Shusen Wang, Zhihua Zhang, Federated Reinforcement Learning with Environment Heterogeneity, PMLR 202- W. Xiong, Q. Liu, F. Li, B. Wang, F.

Zhu, Personalized federated reinforcement learning:

Balancing personalization and experience sharing via distance constraint, Elsevier Expert Systems with Applications, Volume 238, Part F, 202- C.

Zhang, H. Wang, A. Mitra, J. Anderson, Finite-Time Analysis of On-Policy Heterogeneous Federated Reinforcement Learning, ICLR 202- K. Avrachenkov, and V. Borkar, Whittle Index Based Q-learning for Restless Bandits with Average Reward. Automatica, 139, 110186, 202- O. Marfoq, G. Neglia, A. Bellet, L. Kameni, and R. Vidal, Federated Multi-Task Learning under a Mixture of Distributions, NeurIPS, 202- O. Marfoq, L. Kameni, R. Vidal, G. Neglia, Personalized Federated Learning through Local Memorization, ICML, 2022.

Principales activités:

Main activities of a PhD student include conducting research and writing research papers.

Compétences:

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:

Duration: 36 months

Locati