24–29 Nov 2024
University of Melbourne
Australia/Melbourne timezone

Benchmarking Quantum Convolutional Neural Networks for Classification and Data Compression Tasks

Not scheduled
20m
Carillo Gantner Theatre (University of Melbourne)

Carillo Gantner Theatre

University of Melbourne

Sidney Myer Asia Centre, Swanston St, University of Melbourne VIC 3010, Australia
Poster Benchmarking quantum machine learning Posters

Speakers

Dr Chee Kwan Gan (Institute of High Performance Computing, A*STAR) Jian Feng Kong (Institute of High Performance Computing, A*STAR)

Description

Quantum machine learning leverages on quantum parallelism and entanglement to enhance classical machine learning WHITE algorithms performance. Quantum Convolutional Neural Networks (QCNNs) have shown significant potential in classification tasks by exploiting quantum parallelism and entanglement to process quantum data with short range entanglement structure, and avoids the notorious barren plateau problem by construction due to its logarithmic circuit depth. Here we compare the efficiency of QCNN with that of the hardware-efficient ansatz (HEA) in classifying the ground states of two quantum models. We have also studied the compression capabilities of the ground states of one quantum model.

Primary authors

Dr Jun Yong Khoo (Institute of High Performance Computing, A*STAR) Dr Chee Kwan Gan (Institute of High Performance Computing, A*STAR) Dr Wenjun Ding Stefano Carrazza Dr Jun Ye (Institute of High Performance Computing, A*STAR) Jian Feng Kong (Institute of High Performance Computing, A*STAR)

Presentation materials