A Quantum Machine Learning Approach to Content-Based Media Recommendation

Mina Abbaszadeh; Ouissal Moumou; Anna Pearson; Bob Coecke; Rob Cooper; Alessandro Piscopo; Mehrnoosh Sadrzadeh

cs.LGOct 15, 2026

A Quantum Machine Learning Approach to Content-Based Media Recommendation

Authors: Mina Abbaszadeh, Ouissal Moumou, Anna Pearson, Bob Coecke, Rob Cooper, Alessandro Piscopo, Mehrnoosh Sadrzadeh

Organizations: Quantinuum United Kingdom · University College London United Kingdom · Relational Intelligence United Kingdom · British Broadcasting Corporation United Kingdom

Abstract

We use quantum machine learning techniques to develop a content-based media recommender system and evaluate its performance on TV programmes using textual descriptions from one week of BBC iPlayer data. Grounded in the Quantum Natural Language Processing framework (QNLP), programme synopses are mapped to variational quantum circuits using their semantic structure. The parameters of these circuits are trained via a supervised pairwise similarity learning task. We benchmark the quantum recommender against a classical BiLSTM baseline trained on the same dataset and objective. While both models achieve comparable performance on the similarity learning task, the quantum model requires significantly fewer parameters (47k vs. 17.8M). When the BiLSTM is constrained to a matching parameter budget, its performance degrades substantially (by 3.63 percentage points in Mean Average Precision (MAP)), indicating that the structured quantum representation provides a highly parameter-efficient alternative in this setting. We demonstrate the feasibility of executing the trained models on quantum hardware by deploying the circuits on Quantinuum's System Model H2 (and accompanying emulator) using a SWAP-test-based protocol. The results confirm that the similarity signal is preserved under realistic, noisy conditions. Taken together, these findings highlight the potential of variational quantum circuits for recommendation tasks and represent a practical step toward their real-world deployment.

Figures & tables

Figure 1. Basic generators in the DisCoCirc framework and their quantum realisation.

Figure 2. Pipeline for quantum text-similarity computation. Each TV programme synopsis is structurally parsed and mapped to a parameterised quantum circuit, producing output states $\rho_{1}$ and $\rho_{2}$ . Their fidelity $F(\rho_{1},\rho_{2})$ yields the final similarity score $s\in[0,1]$ .

Models	Train	Valid	Test	mean train	mean test	Trainable Parameters
QNLP: discard	98.34%	89.00%	90.90%	98.43% $\pm$ 0.49%	90.66% $\pm$ 2.93%	47.4K
QNLP: post-selection	96.69%	87.87%	88.88%	97.44% $\pm$ 0.7%	91.47% $\pm$ 1.77%	47.4K
Q-BOW	71.79%	58.58%	65.65%	72.59% $\pm$ 6.34%	67.11% $\pm$ 10.90%	46.4k
BiLSTM@17M	92.76%	88.89%	90.91%	95.59% ± 0.66%	91.70% ± 1.87%	17.8M
BiLSTM@47K	53.24%	52.52%	53.54%	53.30% $\pm$ 0.86%	53.30% $\pm$ 3.09%	47.8K

Table 1. Results for the pairwise similarity prediction task. Train/Valid/Test results correspond to the initial fixed-split experiment used for hyperparameter tuning. Mean train and mean test report 5-fold cross-validation performance (mean ± standard deviation) using the selected hyperparameters. QNLP is our main quantum recommender; Q-BOW is a quantum baseline that only considers word meaning; the @ signs of the BiLSTMs are their numbers of parameters. For reference, a majority-class baseline that always predicts the positive class achieves an accuracy of 53.3%

Content-Based Models	MAP%	NDCG%	ILD%	Surprisal	Coverage%
Quantum Models:
QNLP: discard	12.93	25.87	78.84	5.31	100.00
QNLP: post-selection	13.33	26.80	78.28	5.30	100.00
Q-BOW	9.30	20.58	79.31	5.63	100.00
Classical Models:
[email protected]	13.03	26.02	79.10	5.37	100.00

Table 2. Evaluation of accuracy and diversity metrics for 145 TV programs.

Figure 3. Pipeline for evaluating learned similarity within a content-based recommender.

Panel of Figure 4 — Figure 4. Destructive SWAP test. Circuits $A$ and $B$ run forward on disjoint registers. A Bell-basis measurement (CNOT + $H$ ) is applied across the output qubit pairs. Discarded qubits ( $\times$ ) are measured but not used.

Metric	Dissimilar	Similar
Training fidelity (Uhlmann)	$0.44\pm 0.19$	$0.85\pm 0.10$
SWAP test overlap	$0.174\pm 0.042$	$0.368\pm 0.065$

Table 3. Comparison of the classical training fidelity and the SWAP test overlap (noiseless emulator, 1000 shots) on 99 programme pairs (46 dissimilar, 53 similar). There is a strong correlation

r=0.910

between the two, confirming that the similarity signal is preserved in the output-qubit state overlap.

Figure 5. Effect of shot budget on the distribution of SWAP test overlaps. Left: 100 shots ( $\mathrm{SE}\approx 0.10$ ), where the distributions overlap. Right: 1000 shots ( $\mathrm{SE}\approx 0.03$ ), where the distributions are more separated. Since $\mathrm{SE}=1/\sqrt{N}$ , increasing the shot count improves the approximation of the true overlap at the cost of longer runtimes.

Figure 6. Training fidelity (Uhlmann) versus emulator SWAP test overlap (noiseless, 1000 shots). The correlation $r=0.910$ confirms that the similarity signal is recoverable from sampled measurements. Red: dissimilar; green: similar.

Figure 7. Distribution of SWAP test overlaps on the H2 emulator (noiseless, 1000 shots) by label. Left: strip plot showing individual pairs. Right: histogram with group means. Dissimilar pairs cluster around $0.174$ and similar pairs around $0.368$ .

Figure 8. Noisy versus noiseless emulator overlap at 1000 shots ( $r=0.944$ ). Points scatter around the diagonal, indicating that the noise model adds per-pair variance but does not systematically degrade the signal.

Figure 9. Emulator SWAP test overlap (noisy, 1000 shots) versus compiled two-qubit gate depth. For circuits below approximately depth 100, the separation between similar and dissimilar pairs is clear. For the deepest circuits (depth $>120$ ), both classes are compressed toward $0.25$ , the overlap value expected for maximally mixed states.

Configuration	Dissimilar	Similar	$r$	Acc.
Noiseless, 1000 shots	$0.174\pm 0.042$	$0.368\pm 0.065$	$0.910$	$91.9\%$
Noisy, 1000 shots	$0.170\pm 0.041$	$0.358\pm 0.077$	$0.875$	$88.9\%$
Noiseless, 100 shots	$0.151\pm 0.101$	$0.342\pm 0.129$	$0.690$	$79.8\%$

Table 4. Emulator SWAP test results at 1000 and 100 shots. Accuracy is computed at the mapped training threshold (

0.213

for noiseless,

0.207

for noisy,

0.184

for 100 shots), obtained by projecting the Uhlmann fidelity threshold of

0.5

into the overlap space via linear regression.

	H2-2 Hardware	Noisy Emulator
Dissimilar ( $n=9$ )	$0.157\pm 0.042$	$0.158\pm 0.044$
Similar ( $n=11$ )	$0.414\pm 0.058$	$0.395\pm 0.061$
Gap	$0.257$	$0.238$
Cohen’s $d$	$5.11$	$4.48$
$r$ with training fidelity	$0.963$	—

Table 5. Hardware (H2-2) and noisy emulator results for the 20 selected pairs at 1000 shots. The hardware gap (

0.257

) exceeds the emulator prediction (

0.238

Figure 10. H2-2 hardware results (1000 shots, 20 pairs). Left: strip plot showing clear separation between dissimilar (red) and similar (green) pairs. Right: histogram of hardware overlaps ( $\mu_{\text{dis}}=0.157$ , $\mu_{\text{sim}}=0.414$ , gap $=0.257$ ).

Figure 11. H2-2 hardware overlap versus training fidelity (Uhlmann), $r=0.963$ . Circuits trained to have high similarity produce high state overlap on hardware.

Figure 12. H2-2 hardware overlap versus noisy emulator overlap for the same 20 pairs ( $r=0.952$ ). Points near the diagonal confirm the emulator is a reliable predictor of hardware performance.

Figure 13. H2-2 hardware overlap versus compiled two-qubit gate depth. The separation between similar and dissimilar pairs is maintained across all depths, including the deepest circuits (depth 150).

Figure 14. H2-2 hardware overlap versus number of qubits. The separation between similar and dissimilar pairs is maintained across all circuit sizes (12 to 23 qubits). The dashed line shows the maximally mixed baseline (0.25).

Quantum NLP

See all 12 papers

Do Quantum Transformers Help? A Systematic VQC Architecture Comparison on Tabular Benchmarks
Apr 27, 2026Chi-Sheng Chen, En-Jui KuoQuantum Circuit Expressibility Quantum Machine Learning
Beth Israel Deaconess Medical Center & Harvard Medical School Boston, MA, USA · Department of Electrophysics National Yang Ming Chiao Tung University Hsinchu, Taiwan
Parameter-Efficient Quantum NLP for Paraphrase Detection: Performance, Robustness, and Entanglement
Sep 13, 2026Farha Nausheen, Khandakar Ahmed, Farina RiazQuantum NLP Quantum Machine Learning
ISILC, Victoria University, Melbourne, VIC, Australia · CSIRO, Australian International Institute of Higher Education, Sydney, NSW, Australia
Tailor Made Embeddings for Quantum Machine Learning
Jun 24, 2026Aldo Lamarre, Dominik ŠafránekVariational Autoencoders Quantum Machine Learning
Department of Electrical Engineering, KAIST, Daejeon, South Korea · Center for Theoretical Physics of Complex Systems, Institute for Basic Science (IBS), Daejeon - 34126, Korea · Faculty of Mathematics and Physics, Charles University, Ke Karlovu 3, 121 16 Praha 2, Czech Republic

Apr 27, 2026J/K move · Enter open · S save

Do Quantum Transformers Help? A Systematic VQC Architecture Comparison on Tabular Benchmarks

Chi-Sheng Chen, En-Jui Kuo

Beth Israel Deaconess Medical Center & Harvard Medical School Boston, MA, USA · Department of Electrophysics National Yang Ming Chiao Tung University Hsinchu, Taiwan

Quantum Circuit Expressibility Quantum Machine Learning Tabular ML Variational Quantum Circuits

Open paper AskPDF arXiv

Variational quantum circuits (VQCs) are a leading approach to quantum machine learning on near-term devices, yet it remains unclear which circuit architecture yields the best accuracy-parameter trade-off on classical tabular data. We present a systematic empirical comparison of four VQC families -- multi-layer fully-connected (FC-VQC), residual (ResNet-VQC), hybrid quantum-classical transformer (QT), and fully quantum transformer (FQT) -- across five regression and classification benchmarks. Our key findings are: (i)~FC-VQCs achieve 90-96% of the

R^2

of attention-based VQCs while using 40-50% fewer parameters, and consistently outperform equal-capacity MLPs (mean

R^2{=}0.829

vs.\ MLP

_{720}

0.753

on Boston Housing, 3-seed average); (ii)~~FC-VQC's Type~~4 inter-block connectivity provides partial cross-token mixing that approximates the role of attention -- explicit quantum self-attention yields only marginal gains on most datasets while significantly increasing parameter count; (iii)~~expressibility saturates at circuit depth~~

{\approx}\,3

, explaining why shallow VQCs already cover the Hilbert space effectively; (iv)~LayerNorm on the fully quantum transformer improves classification accuracy, suggesting normalization is important when all operations are quantum; (v)~in our noise study on Boston Housing, FQT degrades gracefully under depolarizing noise while QT collapses. All results are validated across three random seeds. These findings provide practical architectural guidance for deploying VQCs on near-term quantum hardware.

Abstract

Quantum NLP

Explore similar work

Do Quantum Transformers Help? A Systematic VQC Architecture Comparison on Tabular Benchmarks

Parameter-Efficient Quantum NLP for Paraphrase Detection: Performance, Robustness, and Entanglement

Tailor Made Embeddings for Quantum Machine Learning