In the design of stochastic models, there is a constant trade-off between model complexity and accuracy. Here we prove that quantum models enable a more favorable trade-off. We present a technique for identifying fundamental upper bounds on the predictive accuracy of dimensionality-constrained classical models. We identify quantum models that surpass this bound by creating an algorithm that learns quantum models given time-series data. We demonstrate that this quantum accuracy advantage is attainable in a present-day noisy quantum device. These results illustrate the immediate relevance of quantum technologies to time-series analysis and offer an instance where their resulting accuracy advantage can be provably established.
In the design of stochastic models, there is a constant trade-off between model complexity and accuracy. Here we prove that quantum models enable a more favorable trade-off. We present a technique for identifying fundamental upper bounds on the predictive accuracy of dimensionality-constrained classical models. We identify quantum models that surpass this bound by creating an algorithm that learns quantum models given time-series data. We demonstrate that this quantum accuracy advantage is attainable in a present-day noisy quantum device. These results illustrate the immediate relevance of quantum technologies to time-series analysis and offer an instance where their resulting accuracy advantage can be provably established.
In modelling complex processes, the potential past data that influence future expectations are immense. Models that track all this data are not only computationally wasteful but also shed little light on what past data most influence the future. There is thus enormous interest in dimensional reduction-finding automated means to reduce the memory dimension of our models while minimizing its impact on its predictive accuracy. Here we construct dimensionally reduced quantum models by machine learning methods that can achieve greater accuracy than provably optimal classical counterparts. We demonstrate this advantage on present-day quantum computing hardware. Our algorithm works directly off classical time-series data and can thus be deployed in real-world settings. These techniques illustrate the immediate relevance of quantum technologies to time-series analysis and offer a rare instance where the resulting quantum advantage can be provably established.
Abstract Many inference scenarios rely on extracting relevant information from known data in order to make future predictions. When the underlying stochastic process satisfies certain assumptions, there is a direct mapping between its exact classical and quantum simulators, with the latter asymptotically using less memory. Here we focus on studying whether such quantum advantage persists when those assumptions are not satisfied, and the model is doomed to have imperfect accuracy. By studying the trade-off between accuracy and memory requirements, we show that quantum models can reach the same accuracy with less memory, or alternatively, better accuracy with the same memory. Finally, we discuss the implications of this result for learning tasks.
Abstract Variational quantum machine learning algorithms have been proposed as promising tools for time series prediction, with the potential to handle complex sequential data more effectively than classical approaches. However, their practical advantage over established classical methods remains uncertain. In this work, we present a comprehensive benchmark study comparing a range of variational quantum algorithms (VQAs) and classical machine learning models for time series forecasting. We evaluate their predictive performance on three chaotic systems across 27 time series prediction tasks of varying complexity, and ensure a fair comparison through extensive hyperparameter optimization. Our results indicate that, in many cases, quantum models struggle to match the accuracy of simple classical counterparts of comparable complexity. Furthermore, we analyze the predictive performance relative to the model complexity and discuss the practical limitations of VQAs for time series forecasting.
This research project embarks on a comprehensive comparative analysis of classical and quantum machine learning models specifically tailored for time series forecasting. By employing a range of classical machine learning models alongside quantum machine learning models, the study aims to predict future data points with heightened accuracy and efficiency. Leveraging AWS Braket, the project facilitates the deployment of quantum models on both quantum simulators and quantum hardware devices, providing a robust platform for empirical evaluation. The goal is to meticulously assess the performance metrics, computational efficiency, and predictive accuracy of these models. Ultimately, the research aspires to enhance the understanding and practical application of quantum machine learning techniques in the realm of time series forecasting, potentially uncovering new paradigms in predictive analytics. Our research indicates that the performance metrics for classical machine learning models yielded superior results compared to those of quantum machine learning models for our use case, although when ensembled with classical models, we get better results.
Quantum computing holds unparalleled potentials to enhance machine learning. However, a demonstration of quantum learning advantage has not been achieved so far. We make a step forward by rigorously establishing a noise-robust, unconditional quantum learning advantage in expressivity, inference speed, and training efficiency, compared to commonly-used classical models. Our proof is information-theoretic and pinpoints the origin of this advantage: entanglement can be used to reduce the communication required by non-local tasks. In particular, we design a task that can be solved with certainty by quantum models with a constant number of parameters using entanglement, whereas commonly-used classical models must scale linearly to achieve a larger-than-exponentially-small accuracy. We show that the quantum model is trainable with constant resources and robust against constant noise. Through numerical and trapped-ion experiments on IonQ Aria, we demonstrate the desired advantage. Our results provide valuable guidance for demonstrating quantum learning advantages with current noisy intermediate-scale devices.
One of the key challenges in quantum machine learning is finding relevant machine learning tasks with a provable quantum advantage. A natural candidate for this is learning unknown Hamiltonian dynamics. Here, we tackle the supervised learning version of this problem, where we are given random examples of the inputs to the dynamics as classical data, paired with the expectation values of some observable after the time evolution, as corresponding output labels. The task is to replicate the corresponding input-output function. We prove that this task can yield provable exponential classical-quantum learning advantages under common complexity assumptions in natural settings. To design our quantum learning algorithms, we introduce a new method, which we term \textit{\subroutine}~algorithm for parametrized circuit functions, and which may be of independent interest. Furthermore, we discuss the limitations of generalizing our method to arbitrary quantum dynamics while maintaining provable guarantees. We explain that significant generalizations are impossible under certain complexity-theoretic assumptions, but nonetheless, we provide a heuristic kernel method, where we trade-off provable correctness for broader applicability.