Learning to Program Variational Quantum Circuits with Fast Weights

TL;DR

This paper introduces Quantum Fast Weight Programmers (QFWP), a hybrid classical-quantum approach that efficiently learns temporal dependencies in quantum machine learning tasks without relying on quantum recurrent neural networks.

Contribution

The paper proposes QFWP, a novel hybrid model that uses a classical neural network to dynamically update quantum circuit parameters, improving training efficiency and temporal learning capabilities.

Findings

QFWP achieves comparable or better performance than QLSTM in time-series prediction.

The model reduces training complexity by avoiding full quantum recurrent neural networks.

Numerical simulations validate the effectiveness of QFWP in RL and time-series tasks.

Abstract

Quantum Machine Learning (QML) has surfaced as a pioneering framework addressing sequential control tasks and time-series modeling. It has demonstrated empirical quantum advantages notably within domains such as Reinforcement Learning (RL) and time-series prediction. A significant advancement lies in Quantum Recurrent Neural Networks (QRNNs), specifically tailored for memory-intensive tasks encompassing partially observable environments and non-linear time-series prediction. Nevertheless, QRNN-based models encounter challenges, notably prolonged training duration stemming from the necessity to compute quantum gradients using backpropagation-through-time (BPTT). This predicament exacerbates when executing the complete model on quantum devices, primarily due to the substantial demand for circuit evaluation arising from the parameter-shift rule. This paper introduces the Quantum Fast Weight Programmers (QFWP) as a solution to the temporal or sequential learning challenge. The QFWP leverages a classical neural network (referred to as the 'slow programmer') functioning as a quantum programmer to swiftly modify the parameters of a variational quantum circuit (termed the 'fast programmer'). Instead of completely overwriting the fast programmer at each time-step, the slow programmer generates parameter changes or updates for the quantum circuit parameters. This approach enables the fast programmer to incorporate past observations or information. Notably, the proposed QFWP model achieves learning of temporal dependencies without necessitating the use of quantum recurrent neural networks. Numerical simulations conducted in this study showcase the efficacy of the proposed QFWP model in both time-series prediction and RL tasks. The model exhibits performance levels either comparable to or surpassing those achieved by QLSTM-based models.