The impressive progress in ultrafast laser technology, ranging from the femtosecond to the attosecond timescale and from the THz to the XUV frequency range, is making possible to probe real-time electronic and nuclear dynamics in atoms, molecules and solids[1]. Fundamental insight can be gained into the primary photoinduced processes in systems with growing level of complexity[2]. The capability of following and steering ultrafast dynamics has tremendous impact in a wide range of applications, from materials science[3] to life sciences.

Clearly, advances in theories and methods inevitably require an intense exchange with the experimental community due to the complexity of the systems and of the measurements. In the last decade the effort in developing predictive and computationally feasible methods has virtually exploded. Ab initio approaches based DFT and nonequilibrium Green’s function (NEGF)[4] have recently made contact with time-resolved experiments in 2D systems and nanostructures. Other ab initio methods based on wavefunctions (e.g., ADCn, CASPTn) or reduced quantities (e.g., TDDFT, NEGF) are opening up high prospects to access the electron-nuclear subfemtosecond dynamics in molecules. Furthermore, accurate real-time numerical methods have been put forward for strongly correlated model systems (e.g., TD- DMFT and DMRG). Gathering world-leading experts in these very different techniques is a unique cross- fertilization opportunity which this workshop will provide to advance the current ab-initio state-of-the-art. In fact, the rapid development of experimental techniques have not been followed by a simultaneous integration with the ab-initio computational community. The result is a scarce availability of numerical tools for crucial systems of technological or fundamental interest, e.g., biological molecules, large nanostructures and materials with strong correlation and/or spin- orbit coupling. One of the key challenge is therefore to extend the range of application of material science and chemistry codes to the study of out-of-equilibrium properties. For this purpose, it is crucial to let experimental, theoretical and computational scientists meet and debate on crucial questions like: how to extend the accuracy of ab- Initio methods out-of-equilibrium? How to efficiently benefit from the advances in computation facilities to simulate the nonequilibrium dynamics of large molecules, nanostructures and solids? How to translate laser-pulse features (pulse center frequency, bandwidth, duration, fluence, polarization) into boundary conditions and suitable approximations for the computational tools? Can we devise a series of tools and procedures to provide to the community?

This workshop aims at being a turning point in ultrafast computational science, settling down crucial and yet unexplored directions to progress. We will confront different theoretical formulations of experimental outcomes, discuss their range of applicability as well as their physical and numerical limitations. We will also discuss for the various approaches how to include the missing physics and whether this inclusion is numerically feasible.

[1] F. Krausz and M. Ivanov, Rev. Mod. Phys. 81, 163 (2009).
[2] M. Först et al, Nature Physics 7, 854–856 (2011)
[3] D. Fausti, et al. Science 331, 189–191 (2011)
[4] G. Stefanucci and R. van Leeuwen, Nonequilibrium Many-Body Theory of Quantum Systems: A Modern Introduction (Cambridge University Press, Cambridge, 2013)