NASA Logo, National Aeronautics and Space Administration

+NASA Home

+Ames Home

Quantum Data Mining Paper Led By QuAIL Intern to be Published in Top Data Sciece Conference Proceedings
Intelligent Systems Division Banner

Quantum Data Mining Paper Led By QuAIL Intern to be Published in Top Data Sciece Conference Proceedings

The Quantum Artificial Intelligence Laboratory (QuAIL) quantum data mining paper, “High Dimensional Similarity Search with Quantum-Assisted Variational Autoencoder”, led by QuAIL intern Nicholas Gao, will be published in the upcoming Knowledge Discovery and Data mining conference (KDD-2020) proceedings.

BACKGROUND: Time series analysis is key to understanding and uncovering changes in the Earth system. The overall goal of the proposed effort is to develop fast and efficient analysis of time series data from NASA’s satellite and derived datasets at large scale (billions of time series from 100’s of terabytes to petabytes of data) that would be deployed on the NASA Earth Exchange (NEX) and be accessible to both supercomputing and quantum computing environments.

As a part of this project we demonstrated the application of a quantum machine learning algorithm to a high-dimensional Earth sciences dataset. We focus on the problem of similarity search in high-dimensional time series. Similarity search is an important problem for a variety of application areas, such as computer vision and recommender systems, and has a rich research history. Recently, some researchers have looked to novel machine learning approaches for improvements over state-of-the-art methods. This new approach to the problem applies a quantum machine learning algorithm and quantum processing (in our case, on the D-Wave 2000q quantum annealer housed at NASA Ames). By adapting existing machine learning algorithms, we can integrate quantum devices into subroutines where they may, in the future, speed up or improve on current methods; in this case, sampling from Boltzmann distributions. Additionally, this approach achieves low memory requirements, which is a key requirement for the problem at hand. Though current quantum processors are small and noisy, we find that in this instance the quantum annealer performs well enough to warrant research and exploration of possible real-world applications.

NASA PROGRAM FUNDING: This research was supported by a grant from NASA’s Advanced Information Systems Technology (AIST) program

TEAM: Nicholas Gao (lead author), Eleanor Rieffel (QuAIL), Ramakrishna Nemani (NEX), Thomas Vandal (NEX), Walter Vinci (QuAIL), and Max Wilson (QuAIL)

POINT OF CONTACT: Eleanor Rieffel, eleanor.rieffel@nasa.gov

First Gov logo
NASA Logo - nasa.gov