[Aspuru-Guzik group list] group meeting tomorrow, 2pm

Hello Everyone, Tomorrow I'll be giving my first group meeting at the usual time, in the usual place (Division Room, 2pm). My plan is to present two projects that I've been working on recently; thus, I'll include two abstracts. Part 1: Effective classical force-fields for quantum ensembles Most approaches to molecular dynamics involve integrating Newton's equations on force-fields which are designed to approximate quantum potential energy surfaces. While this seems justifiable in the limit of high temperature and/or heavy atoms, the "highly quantum" behavior of ensembles significantly affected by low mass particles such as hydrogen cannot be reproduced via classical propagation with such force-fields. However, the success of path integral approaches such as RPMD beg the question of whether there may exist local "effective potentials" which can approximately reproduce quantum ensemble distributions when integrated classically. Using a simple genetic algorithm routine we have been able to find effective potentials which closely reproduce the ensemble position distribution of several simple systems. My presentation of this project will cover the theoretical underpinnings and successful applications of our technique. Part 2: Heuristic approaches to embedding NP Hard problems in Ising models suitable for adiabatic quantum computation A prominent architecture for prototype quantum annealing devices (in particular the Dwave One) is designed to find the ground state of Ising Hamiltonians with restricted couplings. Even if such a device were to function in an idealized fashion there are many NP Hard problems which cannot be mapped to this subset of Ising models at all, which cannot be mapped in an efficient way, or for which finding the mapping is itself an NP Hard problem. In the second part of my group meeting I will discuss techniques that we've used which attempt to find a heuristic embedding to such Hamiltonians. In particular, I will discuss our work studying the optimization of water cluster configurations with DWave One. -- Ryan Babbush | PhD Student in Chemistry (949) 331-3943 | babbush(a)fas.harvard.edu Harvard University | Aspuru-Guzik Research Group 12 Oxford Street, Box 400 | Cambridge, MA 02138