Adaptive Localized Quantum Subspace Methods for Large-Scale Molecular Electronic Structure Simulation
Develop a scalable hybrid quantum–classical framework for electronic-structure simulation of large and strongly correlated molecular systems using sample-based quantum diagonalization and localized active-space techniques.
The goal is not simply to reproduce an existing algorithm, but to build an adaptive, resource-aware quantum chemistry workflow on top of recent state-of-the-art methods, benchmark it against classical approaches, and investigate pathways toward increasingly large molecular simulations.
Tools-Technologies | C, Jupyter Python Notebooks, PowerAI, Spark |
Platform | 1 ) IBM Bluemix www.bluemix.com |
College | All College |
richa goel' Comment
The project aims to answer:
Can localized molecular fragmentation, adaptive quantum sampling, persistent configuration memory, and classical subspace expansion be combined to obtain accurate electronic-structure predictions for increasingly large strongly correlated molecules while minimizing quantum and classical computational resources?