A quantum learning machine for binary classification of qubit states that does not require quantum memory is introduced and shown to perform with the minimum error rate allowed by quantum mechanics for any size of the training set. This result is shown to be robust under (an arbitrary amount of) noise and under (statistical) variations in the composition of the training set, provided it is large enough. This machine can be used an arbitrary number of times without retraining. Its required classical memory grows only logarithmically with the number of training qubits, while its excess risk decreases as the inverse of this number, and twice as fast as the excess risk of an estimate-and-discriminate machine, which estimates the states of the training qubits and classifies the data qubit with a discrimination protocol tailored to the obtained estimates.
A quantum learning machine for binary classification of qubit states that does not require quantum memory is introduced and shown to perform with the minimum error rate allowed by quantum mechanics for any size of the training set. This result is shown to be robust under (an arbitrary amount of) noise and under (statistical) variations in the composition of the training set, provided it is large enough. This machine can be used an arbitrary number of times without retraining. Its required classical memory grows only logarithmically with the number of training qubits, while its excess risk decreases as the inverse of this number, and twice as fast as the excess risk of an estimate-and-discriminate machine, which estimates the states of the training qubits and classifies the data qubit with a discrimination protocol tailored to the obtained estimates.
A quantum learning machine for binary classification of qubit states that does not require quantum memory is introduced and shown to perform with the minimum error rate allowed by quantum mechanics for any size of the training set. This result is shown to be robust under (an arbitrary amount of) noise and under (statistical) variations in the composition of the training set, provided it is large enough. This machine can be used an arbitrary number of times without retraining. Its required classical memory grows only logarithmically with the number of training qubits, while its excess risk decreases as the inverse of this number, and twice as fast as the excess risk of an "estimate-and-discriminate" machine, which estimates the states of the training qubits and classifies the data qubit with a discrimination protocol tailored to the obtained estimates.
Pattern recognition is a central topic in learning theory, with numerous applications such as voice and text recognition, image analysis and computer diagnosis. The statistical setup in classification is the following: we are given an i.i.d. training set ( X 1 , Y 1 ), ... , ( X n , Y n ), where X i represents a feature and Y i ∊{0, 1} is a label attached to that feature. The underlying joint distribution of ( X , Y ) is unknown, but we can learn about it from the training set, and we aim at devising low error classifiers f: X → Y used to predict the label of new incoming features. In this paper, we solve a quantum analogue of this problem, namely the classification of two arbitrary unknown mixed qubit states. Given a number of 'training' copies from each of the states, we would like to 'learn' about them by performing a measurement on the training set. The outcome is then used to design measurements for the classification of future systems with unknown labels. We found the asymptotically optimal classification strategy and show that typically it performs strictly better than a plug-in strategy, which consists of estimating the states separately and then discriminating between them using the Helstrom measurement. The figure of merit is given by the excess risk equal to the difference between the probability of error and the probability of error of the optimal measurement for known states. We show that the excess risk scales as n −1 and compute the exact constant of the rate.
To witness quantum advantages in practical settings, substantial efforts are required not only at the hardware level but also on theoretical research to reduce the computational cost of a given protocol. Quantum computation has the potential to significantly enhance existing classical machine learning methods, and several quantum algorithms for binary classification based on the kernel method have been proposed. These algorithms rely on estimating an expectation value, which in turn requires an expensive quantum data encoding procedure to be repeated many times. In this work, we calculate explicitly the number of repetition necessary for acquiring a fixed success probability and show that the Hadamard-test and the swap-test circuits achieve the optimal variance in terms of the quantum circuit parameters. The variance, and hence the number of repetition, can be further reduced only via optimization over data-related parameters. We also show that the kernel-based binary classification can be performed with a single-qubit measurement regardless of the number and the dimension of the data. Finally, we show that for a number of relevant noise models the classification can be performed reliably without quantum error correction. Our findings are useful for designing quantum classification experiments under limited resources, which is the common challenge in the noisy intermediate-scale quantum era.
The primary question in the emerging field of machine learning (QML) is: can we achieve speed-up of machine learning? and/or what and how the effects contribute to the speed-up? The positive answers to such questions have recently been provided, for example, by the support vector machine (QSVM) which classifies many data vectors exponentially fast. However, to realize such an exponential learning speed-up, it is demanded to develop full-scale computer with a specific device, so-called quantum random access memory (QRAM)---which stores the classical big data, extremely fast, in the novel form of being quantum-mechanically superposed. Therefore, another question now arises: is it possible to attain learning speed-up even with the classical training data, and hence, requiring no full-scale computer and QRAM? To seek a satisfactory answer to this question, we consider a binary classification learning where the classicality of the training data are preserved. We then focus on a small unit structure for whole classification task, in which the minimum achievable learning speed-up would be exhibited. We provide the analysis and proof-of-principle experimental validation of a certain degree of learning speed-up in such a small unit, advocating an affirmative answer to the aforementioned main question. Through further investigation, it is also shown that our QML scheme is quite robust against the decoherence.
We consider the problem of correctly classifying a given quantum two-level system (qubit) which is known to be in one of two equally probable quantum states. We assume that this task should be performed by a quantum machine which does not have at its disposal a complete classical description of the two template states, but can only have partial prior information about their level of purity and mutual overlap. Moreover, similarly to the classical supervised learning paradigm, we assume that the machine can be trained by n qubits prepared in the first template state and by n more qubits prepared in the second template state. In this situation, we are interested in the optimal process which correctly classifies the input qubit with the largest probability allowed by quantum mechanics. The problem is studied in its full generality for a number of different prior information scenarios and for an arbitrary size n of the training data. Finite size corrections around the asymptotic limit n → ∞ are derived. When the states are assumed to be pure, with known overlap, the problem is also solved in the case of d-level systems.
Quantum machine learning for classical data is currently perceived to have a scalability problem due to (i) a bottleneck at the point of loading data into quantum states, (ii) the lack of clarity around good optimization strategies, and (iii) barren plateaus that occur when the model parameters are randomly initialized. In this work, we propose techniques to address all of these issues. First, we present a quantum classifier that encodes both the input and the output as binary strings which results in a model that has no restrictions on expressivity over the encoded data but requires fast classical compression of typical high-dimensional datasets to only the most predictive degrees of freedom. Second, we show that if one parameter is updated at a time, quantum models can be trained without using a classical optimizer in a way that guarantees convergence to a local minimum, something not possible for classical deep learning models. Third, we propose a parameter initialization strategy called sub-net initialization to avoid barren plateaus where smaller models, trained on more compactly encoded data with fewer qubits, are used to initialize models that utilize more qubits. Along with theoretical arguments on efficacy, we demonstrate the combined performance of these methods on subsets of the MNIST dataset for models with an all-to-all connected architecture that use up to 16 qubits in simulation. This allows us to conclude that the loss function consistently decreases as the capability of the model, measured by the number of parameters and qubits, increases, and this behavior is maintained for datasets of varying complexity. Together, these techniques offer a coherent framework for scalable quantum machine learning.