Dec 1 – 3, 2021
America/New_York timezone

Quantum Classifiers Hybrids with Advanced Data Compression Methods for Higgs Identification

Dec 2, 2021, 9:50 AM
20m

Speaker

Vasilis Belis (ETH Zurich (CH))

Description

We study the quantum counterpart of Support Vector Machines, namely Quantum Support Vector Machines (QSVΜ), and a Quantum Machine Learning (QML) architecture that combines a classical encoder neural network and a Variation Quantum Circuit (VQC) into a single model. That is, a Neural Network Variational Quantum Circuit (NNVQC) for the binary classification of High Energy Physics data. Specifically, we focus on the identification of the Higgs boson in the $t\bar{t}H(b\bar{b})$ channel. Quantum computing approaches can potentially tackle this computationally expensive task by leveraging the so-called quantum feature maps to encode classical data into quantum states. Recent proposals based on the kernel trick assume a one-feature-to-one-qubit mapping of the data. The limited number of available qubits on Noisy Intermediate-Scale Quantum (NISQ) devices imposes the need for feature compression on complex datasets. The challenge is to maintain sufficient information to achieve a high classification accuracy while performing an effective reduction.

This contribution assesses the effect of different data compression and dimensionality reduction techniques with respect to quantum machine learning algorithms. We develop five distinct Auto-Encoder architectures, including a Variational and an end-to-end Sinkhorn Auto-Encoder with a classical classification neural network attached to its latent space. The latent spaces produced with optimal hyperparameters and data normalisation were passed to a QSVM that was used to perform the $t\bar{t}H(b\bar{b})$ classification. The QSVM performance is improved for some of the considered Auto-Encoder latent spaces. The classification power of the NNVQC and of its classical counterparts are comparable.

The training and performance of the quantum models is affected by noise inherent to NISQ devices. We investigated the influence of different types of quantum hardware noise and we concluded that the tested QML models are suitable for operation on current NISQ devices.

Primary authors

Presentation materials