Abstract: When cooled fast enough to avoid crystallization, a liquid becomes increasingly viscous and eventually forms a glass. This glass transition, one of the oldest unsolved problems in condensed-matter physics, has given rise to a wide diversity of views. The mean-field theory provides an elaborate and consistent picture of glasses and glass formation that is exact for liquids in the limit of infinite spatial dimensions. What, however, remains of this description in finite dimensions where spatial fluctuations generate mechanisms of all kinds that may alter the picture? I will review the issue and discuss possible strategies to make progress.

Prof. Gilles TARJUS is the Head of the Statistical Physics group at the CNRS, LPTMC (Dept. of Condensed-Matter Theoretical Physics), Sorbonne University, Paris (France). Prof. Gilles TARJUS was appointed as a Research Charge at the CNRS, LPTMC in 1979, and then promoted to the Research Director in Paris in 1994. Before it, he worked as a Research fellow of Chemistry and Biochemistry, at UCLA, California (1987-1989) and was designated as an invited professor in 1996. He earned his doctoral degree in physics at the University of Pierre & Marie Curie (Paris VI) in 1980 and a diploma from Ecole Polytechnique (Parisi) in 1977. His work in the field of statistical physics and condensed-matter theory has been focused for the last 20 years on glassy and disordered systems. He has published more than 240 papers in peer-reviewed journals and has given 160 invited talks and seminars. He has been the advisor or co-advisor of 13 Ph.D. students and 8 post-docs.