Integrated Calendar

A new quantum theory, MRSF-TDDFT (Mixed-Reference Spin-Flip Time-Dependent Density Functional Theory) has been developed*, which introduces the multi-reference advantages within the linear response formalism. The density functional theory (DFT) and linear response (LR) time dependent (TD)-DFT are of utmost importance for routine computations. However, the single reference formulation of DFT is suffering from the description of open-shell singlet systems such as diradicals and bond-breaking. LR-TDDFT, on the other hand, finds difficulties in the modeling of conical intersections, doubly excited states, and core-level excitations. Many of these limitations can be overcome by MRSF-TDDFT, providing an alternative yet accurate route for such challenging situations. Now the theory is combined with NAMD, QM/MM, Spin-Orbit Couplings, and Extended Koopman Theorem. Here, we highlight its performances by presenting our recent results by MRSF-TDDFT especially focusing on nonadiabatic molecular dynamics.

In this talk I will present a framework for designing worst-case to average-case reductions. Focusing on the problem of Matrix Multiplication, I will describe a transformation that takes any weak algorithm that is only correct on a small fraction of the inputs, and converts it into an algorithm that is correct on all inputs, while paying only a small overhead in running time.

The talk is based on joint work with Vahid Asadi, Sasha Golovnev, Tom Gur, and Sathyawageeswar Subramanian.