QuantumReservoirPy: A Software Package for Time Series Prediction

TL;DR

The paper addresses the lack of a common framework for quantum reservoir computing by introducing QuantumReservoirPy, a class-based library that unifies quantum reservoir architectures under a consistent API. It defines a formal RC-like workflow for quantum circuits, described by updates $u_{t+1}=f(u_t,x_t)$ and outputs $y_t^{pred}=W_{out} h(u_t)$ with a suitable loss, and demonstrates how QRC can be implemented via modular, testable components. The package provides static, incremental, and custom reservoir implementations, built-in processing pipelines, and seamless integration with Qiskit backends and scikit-learn estimators, alongside documentation and distribution through PyPI and GitHub. This framework enables straightforward comparison, rapid experimentation, and easier adoption of QRC methods for time-series prediction in real-world settings.

Abstract

In recent times, quantum reservoir computing has emerged as a potential resource for time series prediction. Hence, there is a need for a flexible framework to test quantum circuits as nonlinear dynamical systems. We have developed a software package to allow for quantum reservoirs to fit a common structure, similar to that of reservoirpy which is advertised as "a python tool designed to easily define, train and use (classical) reservoir computing architectures". Our package results in simplified development and logical methods of comparison between quantum reservoir architectures. Examples are provided to demonstrate the resulting simplicity of executing quantum reservoir computing using our software package.

Figures (4)

• Figure 1: A quantum reservoir system consists of a learning task, an en- and de-coder (red) and the dynamic system itself (green). In standard RC the machine learning part is linear regression.
• Figure 2: Quantum circuit construction may be customized through the before, during, and after methods and a timeseries processed with the run and predict methods.
• Figure 3: The intended functionality of the run and predict method. The observed input sequence is $\{x_t\}$ and the target sequence $\{y_t\}$. The reservoir $f$ performs an evolution in time.
• Figure 4: A simple example of the usage of QuantumReservoirPy for time series prediction.