TeD-Q: a tensor network enhanced distributed hybrid quantum machine learning framework

TL;DR

TeD-Q is an open-source framework that combines classical and quantum machine learning techniques, enabling efficient simulation, training, and visualization of quantum models with tensor networks and auto-differentiation.

Contribution

It introduces a comprehensive software framework integrating tensor network simulation, auto-differentiation, and visualization for quantum machine learning, enhancing capabilities beyond existing tools.

Findings

Supports large qubit quantum circuit simulation

Enables real-time visualization of quantum training

Provides multiple gradient computation methods

Abstract

TeD-Q is an open-source software framework for quantum machine learning, variational quantum algorithm (VQA), and simulation of quantum computing. It seamlessly integrates classical machine learning libraries with quantum simulators, giving users the ability to leverage the power of classical machine learning while training quantum machine learning models. TeD-Q supports auto-differentiation that provides backpropagation, parameters shift, and finite difference methods to obtain gradients. With tensor contraction, simulation of quantum circuits with large number of qubits is possible. TeD-Q also provides a graphical mode in which the quantum circuit and the training progress can be visualized in real-time.