Machine learning enhances quantum simulations for complex biological systems

Researchers have developed a novel machine-learned approach, QSCI-RBM, to generate compact subspaces for quantum selected configuration interaction (QSCI) within the density matrix embedding theory (DMET) framework. This method utilizes Restricted Boltzmann Machines (RBMs) to learn the distribution of dominant determinants, enabling the targeted generation of high-probability configurations. Applied to a protein-ligand complex, DMET-QSCI-RBM achieved chemical accuracy with only 4% of the configuration subspace, significantly outperforming standard DMET-SQD simulations which failed to reach accuracy even with 20% of the subspace. This technique promises to reduce classical computational overhead and facilitate scalable quantum embedding simulations for complex biological systems. AI

IMPACT This machine learning approach significantly reduces computational cost for quantum simulations, potentially accelerating drug discovery and materials science.

RANK_REASON The cluster contains a research paper detailing a new computational method for quantum physics simulations. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Ashish Kumar Patra
• Carmofur
• Density Matrix Embedding Theory
• DMET-SQD
• QSCI-RBM
• Quantum Diagonalization
• Quantum Selected Configuration Interaction
• Restricted Boltzmann Machines
• SARS-CoV-2

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →