Efficient similarity estimation between quantum states remains a key bottleneck in quantum machine learning. While quantum fidelity provides a natural similarity measure, its evaluation typically relies on pairwise Swap Test circuits, leading to quadratic scaling with batch size. We propose Resource-efficient Quantum Contrastive Learning (ReQCL), a quantum-native framework that leverages coherent superposition to estimate global similarity. By preparing a batch-level superposition and extracting a batch-averaged fidelity from a single measurement, ReQCL avoids explicit pairwise evaluation and reduces the complexity from quadratic to linear in the batch size. We further introduce a fidelity-ratio objective that contrasts individual alignments against a global similarity baseline, enabling effective optimization of quantum representations. Numerical simulations suggest that the proposed approach learns meaningful embeddings in quantum state space. Our results show that global fidelity estimation via quantum superposition provides a scalable pathway for quantum representation learning.
Efficient similarity estimation between quantum states remains a key bottleneck in quantum machine learning. While quantum fidelity provides a natural similarity measure, its evaluation typically relies on pairwise Swap Test circuits, leading to quadratic scaling with batch size. We propose Resource-efficient Quantum Contrastive Learning (ReQCL), a quantum-native framework that leverages coherent superposition to estimate global similarity. By preparing a batch-level superposition and extracting a batch-averaged fidelity from a single measurement, ReQCL avoids explicit pairwise evaluation and reduces the complexity from quadratic to linear in the batch size. We further introduce a fidelity-ratio objective that contrasts individual alignments against a global similarity baseline, enabling effective optimization of quantum representations. Numerical simulations suggest that the proposed approach learns meaningful embeddings in quantum state space. Our results show that global fidelity estimation via quantum superposition provides a scalable pathway for quantum representation learning.
Quantum machine learning seeks a computational advantage in data processing by evaluating functions of quantum states, such as their similarity, that can be classically intractable to compute. For quantum advantage to be possible, however, it is essential to bypass costly characterisation of individual data instances in favour of efficient, direct similarity evaluation. Here we demonstrate a sample-optimal, hardware-efficient protocol for estimating quantum similarity -- the state overlap -- using bosonic quantum interference. The sample complexity of this approach is independent of the system dimension and is information-theoretically optimal up to a constant factor. Experimentally, we implement the scheme on \emph{Prakash-1}, a quantum computing platform based on a fully programmable integrated photonic processor. By preparing and interfering qudit states on the chip to directly extract their overlap, we demonstrate classification and online learning of quantum data with high accuracy in realistic noisy experiments. Our results establish joint overlap measurements as a scalable pathway to efficient quantum data analysis and a practical building block for network-integrated quantum machine learning.
Quantum contrastive, metric, and self-supervised learning often expose encoded quantum states to the learner through transition probabilities, especially fidelity. Quantum states are known to possess higher-order relational invariants, but their consequences for learned representations remain unclear. Here we show that a transition-probability-only learning interface can possess exact continuous blind directions in certain quantum-state families. We recover this missing information with a batch operator, built from coherent overlap amplitudes and negative masking, where its second moment q− retains four-state interference. Moreover, q− is directly measurable through two-copy interference and can enter variational learning via methods like parameter shift. In relational quartets derived from toric-code and double-semion states, this fidelity-blind signal encodes inequivalent modular data despite identical pairwise fidelities. Finally, in a four-photon benchmark with preparation drift, augmenting all six pairwise fidelities at two orthogonal probes with the corresponding normalized q− reduces the mean out-of-distribution phase error by 86% at equal total shot budget. These results establish multistate relational observables as measurable, trainable, and physically consequential signals for quantum representation learning beyond pairwise fidelity.
Over the last decade, representation learning, which embeds complex information extracted from large amounts of data into dense vector spaces, has emerged as a key technique in machine learning. Among other applications, it has been a key building block for large language models and advanced computer vision systems based on contrastive learning. A core component of representation learning systems is the projection head, which maps the original embeddings into different, often compressed spaces, while preserving the similarity relationship between vectors. In this paper, we propose a quantum-inspired projection head that includes a corresponding quantum-inspired similarity metric. Specifically, we map classical embeddings onto quantum states in Hilbert space and introduce a quantum circuit-based projection head to reduce embedding dimensionality. To evaluate the effectiveness of this approach, we extended the BERT language model by integrating our projection head for embedding compression. We compared the performance of embeddings, which were compressed using our quantum-inspired projection head, with those compressed using a classical projection head on information retrieval tasks using the TREC 2019 and TREC 2020 Deep Learning benchmarks. The results demonstrate that our quantum-inspired method achieves competitive performance relative to the classical method while utilizing 32 times fewer parameters. Furthermore, when trained from scratch, it notably excels, particularly on smaller datasets. This work not only highlights the effectiveness of the quantum-inspired approach but also emphasizes the utility of efficient, ad hoc low-entanglement circuit simulations within neural networks as a powerful quantum-inspired technique.
Quantum machine learning (QML) has attracted growing interest with the rapid parallel advances in large-scale classical machine learning and quantum technologies. Similar to classical machine learning, QML models also face challenges arising from the scarcity of labeled data, particularly as their scale and complexity increase. Here, we introduce self-supervised pretraining of quantum representations that reduces reliance on labeled data by learning invariances from unlabeled examples. We implement this paradigm on a programmable trapped-ion quantum computer, encoding images as quantum states. In situ contrastive pretraining on hardware yields a representation that, when fine-tuned, classifies image families with higher mean test accuracy and lower run-to-run variability than models trained from random initialization. Performance improvement is especially significant in regimes with limited labeled training data. We show that the learned invariances generalize beyond the pretraining image samples. Unlike prior work, our pipeline derives similarity from measured quantum overlaps and executes all training and classification stages on hardware. These results establish a label-efficient route to quantum representation learning, with direct relevance to quantum-native datasets and a clear path to larger classical inputs.
Convolution operations are foundational to classical image processing and modern deep learning architectures, yet their extension into the quantum domain has remained algorithmically and physically costly due to inefficient data encoding and prohibitive circuit complexity. In this work, we present a resource-efficient quantum algorithm that reformulates the convolution product as a structured matrix multiplication via a novel sparse reshaping formalism. Leveraging the observation that localized convolutions can be encoded as doubly block-Toeplitz matrix multiplications, we construct a quantum framework wherein sparse input patches are prepared using optimized key-value QRAM state encoding, while convolutional filters are represented as quantum states in superposition. The convolution outputs are computed through inner product estimation using a low-depth SWAP test circuit, which yields probabilistic amplitude information with reduced sampling overhead. Our architecture supports batched convolution across multiple filters using a generalized SWAP circuit. Compared to prior quantum convolutional approaches, our method eliminates redundant preparation costs, scales logarithmically with input size under sparsity, and enables direct integration into hybrid quantum-classical machine learning pipelines. This work provides a scalable and physically realizable pathway toward quantum-enhanced feature extraction, opening up new possibilities for quantum convolutional neural networks and data-driven quantum inference.
We present QuCoWE a framework that learns quantumnative word embeddings by training shallow hardwareefficient parameterized quantum circuits PQCs with a contrastive skipgram objective Words are encoded by datareuploading circuits with controlled ring entanglement similarity is computed via quantum state fidelity and passed through a logitfidelity head that aligns scores with the shiftedPMI scale of SGNSNoiseContrastive Estimation To maintain trainability we introduce an entanglementbudget regularizer based on singlequbit purity that mitigates barren plateaus On Text8 and WikiText2 QuCoWE attains competitive intrinsic WordSim353 SimLex999 and extrinsic SST2 TREC6 performance versus 50100d classical baselines while using fewer learned parameters per token All experiments are run in classical simulation we analyze depolarizingreadout noise and include errormitigation hooks zeronoise extrapolation randomized compiling to facilitate hardware deployment