Experimental Quantum Embedding for Machine Learning

Abstract

The classification of big data usually requires a mapping onto new data clusters which can then be processed by machine learning algorithms by means of more efficient and feasible linear separators. Recently, Lloyd et al. have advanced the proposal to embed classical data into quantum ones: these live in the more complex Hilbert space where they can get split into linearly separable clusters. Here, these ideas are implemented by engineering two different experimental platforms, based on quantum optics and ultra‐cold atoms, respectively, where we adapt and numerically optimize the quantum embedding protocol by deep learning methods, and test it for some trial classical data. A similar analysis is also performed on the Rigetti superconducting quantum computer. Therefore, it is found that the quantum embedding approach successfully works also at the experimental level and, in particular, we show how different platforms could work in a complementary fashion to achieve this task. These studies might pave the way for future investigations on quantum machine learning techniques especially based on hybrid quantum technologies.

Quantum Neural Networks for a Supply Chain Logistics Application

“…There has been significant recent development in quantum algorithms for neural networks-so-called quantum neural networks (QNNs). This includes the theoretical [6] and experimental study [7,8] of how to best embed the data into the quantum circuit and maximize the separation of input data in Hilbert space. It also includes efforts to improve the training of QNNs, especially to avoid barren plateaus of parameterized circuits.…”

Beyond Bits: A Review of Quantum Embedding Techniques for Efficient Information Processing

“…Similarly, the studies delineated in [50] and [68] are centered on the utilization of angle encoding techniques for quantum classifiers. Furthermore, a broader review of data encoding methods, particularly within the quantum machine learning (QML) domain, can be found in [38], [41], [49], [69], whereas the research presented in [42] and [43] delve into encoding modalities relevant to quantum error correction and NISQ systems. Although the contributions of M. Weigold et al [45], [46] and the research collectives under S. Lloyd et al [48] have significantly enriched our understanding of data encoding patterns, comprehensive analyses addressing runtime complexity or scalability of all pertinent encoding patterns remain sparse.…”

Multiparameter estimation of continuous-time quantum walk Hamiltonians through machine learning

“…[4][5][6] ML techniques have been proven to be of substantial aid when adapted to quantum states characterization, [7][8][9][10][11][12][13][14] optimization of control strategies, [15][16][17][18] quantum state transport 19,20 as well as for parameter estimation and classification tasks. [21][22][23][24][25][26][27][28] Concerning the characterization of quantum processes, Hamiltonian learning strategies have been extensively investigated in order to provide a reliable solution to this challenge. [29][30][31][32][33][34] Moreover, it is pivotal that the required information is often not directly accessible and must be inferred starting from experimental quantities.…”