Homepage Inria website
  • Inria login
  • The Inria's Research Teams produce an annual Activity Report presenting their activities and their results of the year. These reports include the team members, the scientific program, the software developed by the team and the new results of the year. The report also describes the grants, contracts and the activities of dissemination and teaching. Finally, the report gives the list of publications of the year.

  • Legal notice
  • Cookie management
  • Personal data
  • Cookies

Section: Partnerships and Cooperations

International Initiatives

Lifelong Learning Machines program (DARPA) — STELLAR project

Title: STELLAR (Super Turing Evolving Lifelong Learning ARchitecture)

Coordinator: HRL laboratory (Malibu, USA)

Coordinator for Inria: Jean-Baptiste Mouret

Partners: Stanford University (USA), University of California Irvine (USA), University of Texas Austin (USA), IT University of Copenhagen (Denmark), Loughborough University (United Kingdom), Inria – Nancy Grand Est

Objective: Develop a general-purpose neural super Turing machine for lifelong learning and demonstrate supra-human performance in a simulated autonomous driving context. Our Super Turing Evolving Lifelong Learning ARchitecture (STELLAR) system will power a self-driving agent that continually improves its performance and updates its knowledge unsupervised, rapidly adapts to unforeseen contexts, and learns and consolidates new tasks without forgetting old ones. The project involves deep world models, neuroevolution, quality diversity algorithms, and plastic neural networks.

Informal International Partners
  • Oxford University (Shimon Whiteson): data-efficient robot learning[22]

  • Union College (John Rieffel): resilient tensegrity robots [10]

  • Italian Institute of Technology (Enrico Mingo-Hoffman, Daniele Pucci, Nikos Tsagarakis): whole-body control of humanoids [11], [24], [27]

  • IT University Copenhagen (Sebastian Risi): quality diversity algorithms

  • Imperial College (Antoine Cully): data-efficient learning and quality diversity

  • Hochschule Bonn-Rhein-Sieg (Alexander Asteroth): surrogate modelling [17], [7]

  • Kyushu Institute of Technology, Japan (Sozo Inoue, Moe Matsuki): activity recognition