2021-03965 - Post-Doctoral Research Visit F/M Information and Decision Making

Level of qualifications required : PhD or equivalent

Fonction : Post-Doctoral Research Visit

About the research centre or Inria department

The Inria Sophia Antipolis - Méditerranée center counts 34 research teams as well as 8 support departments. The center's staff (about 500 people including 320 Inria employees) is made up of scientists of different nationalities (250 foreigners of 50 nationalities), engineers, technicians and administrative staff. 1/3 of the staff are civil servants, the others are contractual agents. The majority of the center’s research teams are located in Sophia Antipolis and Nice in the Alpes-Maritimes. Four teams are based in Montpellier and two teams are hosted in Bologna in Italy and Athens. The Center is a founding member of Université Côte d'Azur and partner of the I-site MUSE supported by the University of Montpellier.


This postdoctoral position is  hosted at INRIA, Sophia Antipolis, in the frame of the Exploratory Action "Information and Decision Making (IDEM)". It develops within a close collaboration between INRIA, the University of Sheffield (TUOS) and Princeton University in the broad intersection of information theory, game theory, and artificial intelligence. Mobility between INRIA, the Department of Automatic Control and Systems Engineering at The University of Sheffield, as well as the department of Electrical Engineering at Princeton University, is expected depending on the evolution of the COVID-19 pandemic. The position is supervised by Samir M. Perlaza and Alain Jean-Marie.



In most decentralized decision making processes, decisions are made with partial information about the stochastic phenomena underlying the environment in which decision-makers interact. Therefore, information provisioning processes play a central role consisting in transmitting signals to the decision makers aiming at steering the decision making process and seeking for particular outcomes. That is, information can be provided to an individual for increasing its benefit, but also, information can be withhold or distorted for penalizing individuals. Either way, this vulnerability to malicious or undesired influence represents a thread for any decentralized decision making process and thus, it must be understood and properly modeled.

This project aims to study the fundamental limits on the influence that decision makers involved on a common decision making process can exert on each other via revealing, hiding or distorting information. The focus is on the case in which the benefit obtained by an individual decision maker depends upon the decisions of all involved individuals. This situation arises in most decision making processes involving humans, machines or humans and machines: (a) decentralized machine learning; (b) Marketing policies that propose goods to potential custumers; (c) Geo-localization-based navigation applications that provide advise to drivers; and (d) Stock traders that follow different sources of information to buy, sell and trade shares.


Main activities

The postdoc is expected to collaborate with the Inria team NEO to characterize the interplay between data acquisition and information processing in decentralized decision making by bringing together tools from information theory and game theory. This characterization is central in the comprehension of problems including decentralized optimization and Machine Learning subject to local information constraints. The expected research outputs are the fundamental limits of three performance metrics:

(i) Influence is a measure of the deviation of individuals from the rational behavior, in the sense of optimizing a utility, that would have been adopted if influences are not present. Individual behaviors can be modeled by probability measures over the set of decisions. Thus, any measure of distance, e.g., total variation, or pseudo-distance, e.g., relative entropy, between probability measures is a candidate for measuring influence. Adopting one of these measures depends on the generality of the theory that can be derived.

(ii) Feasible outcomes of the interaction between decision makers are often observed with respect to a stability criteria. These criteria are often equilibrium concepts brought from game theory such as Nash equilibria, Stackelberg equilibria, and satisfaction equilibria, among others, which are generally benchmarked against some sort of “social optimum”. Regardless of the equilibrium concept used as reference, the final outcome of the interaction depends upon the information (recommendations) used by the decision makers. Hence, two sets of outcomes are identified: (a) The set of outcomes that can be observed as equilibria (achievable outcomes); and (b) The set of outcomes that will never be observed at an equilibrium given the information exchanges among the decision makers.

(iii) Information rates at which information can be sent over physical channels to the decision makers, including its data acquisition and information processing constraints. That is, the minimum rates needed for influencing the decision makers and lead to a particular outcome.


Candidates are expected to have a PhD degree in mathematics or areas related to information theory and game theory. Previous knowledge on information theory, and game theory is desirable. The candidate must have a provable level of written and spoken english. Skills in french language are not required.

Benefits package

  • 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


Gross Salary: 2653 € per month