Workshops Digital Media Quantum Computing for Media Recommendation

Digital Media Full Day Workshop

Quantum Computing for Media Recommendation and Audience Personalisation

A technical workshop for data science and product teams at streaming platforms and digital publishers assessing where quantum machine learning fits within recommendation and personalisation pipelines, and where it does not.

Full day (6 hours + Q&A)

In person or online

Max 30 delegates

Commission This Workshop View Agenda

Proud to recommend our expert members

Workshop Description

Recommendation engines at streaming scale process billions of user-item interactions daily. Classical approaches (matrix factorisation, deep collaborative filtering, transformer-based sequential models) are mature and performant. The question for data science teams is whether quantum computing offers a meaningful capability that classical methods cannot replicate, and if so, on what timeline.

This workshop works through the specific quantum ML techniques proposed for recommendation: quantum kernel estimation for content similarity, variational quantum circuits for collaborative filtering, and quantum-enhanced embedding for large content catalogues. For each technique, we examine the published results, identify where dequantisation (Tang 2019 and subsequent work) eliminates the claimed quantum speedup, and assess what problem structures would need to hold for quantum advantage to survive. Current NISQ hardware limitations are assessed honestly: circuit depth, qubit coherence times, and gate fidelities against the scale of production recommendation workloads. Participants leave with a framework for evaluating quantum ML vendor claims and a realistic timeline for when quantum recommendation becomes commercially relevant.

What participants cover

Quantum kernel estimation (QKE) for content catalogue embedding: mapping item features to quantum feature spaces and comparing similarity computation against classical RBF kernels
Variational quantum circuit (VQC) architectures for collaborative filtering: parameterised ansatz design, barren plateau mitigation, and hybrid quantum-classical training loops
Dequantisation analysis: understanding which claimed quantum speedups for recommendation have been matched by classical algorithms and which remain open
NISQ hardware honest assessment: current qubit counts, gate fidelities, and coherence times versus the circuit requirements of production-scale recommendation
Quantum-inspired classical methods (tensor networks, random feature sampling) that deliver near-term value without quantum hardware
Vendor claim evaluation framework: structured methodology for assessing quantum ML proposals against classical baselines on your own datasets

Preliminary Agenda

Full-day session structure with scheduled breaks. Content is configurable to your recommendation stack, data scale, and ML infrastructure.

#	Session	Topics
1	Classical Recommendation at Scale Where quantum approaches enter the recommendation pipeline
2	Quantum Kernel Methods for Content Similarity Quantum kernel estimation and feature maps for catalogue embedding	Quantum kernel estimation (QKE): mapping content features to high-dimensional Hilbert space for similarity computation Quantum feature maps versus classical RBF and polynomial kernels: theoretical expressivity and the role of data structure Practical constraints: current QKE implementations on IBM Eagle (127 qubits) require circuit depths that exceed coherence times for production-scale catalogues
Break, after 50 min
3	Variational Quantum Circuits for Collaborative Filtering VQC architectures for user preference modelling	Variational quantum circuit (VQC) design for collaborative filtering: parameterised ansatz, entanglement structure, and the barren plateau problem Hybrid quantum-classical training loops: quantum circuit as feature extractor feeding classical recommendation heads Published results: Netflix-scale matrix factorisation remains beyond NISQ capacity; sub-problems (cold-start user embedding) show more tractable circuit requirements
4	Interactive Demonstration Facilitator-led quantum recommendation pipeline walkthrough	Live demonstration: quantum kernel similarity computation on a sample content catalogue using Qiskit Runtime on IBM Quantum Interpreting kernel matrices: comparing quantum versus classical similarity rankings on the same dataset Discussion: identifying which stages of your recommendation pipeline could benefit from quantum feature maps versus where classical methods remain superior
Break, after 60 min
5	Dequantisation and Classical Benchmarking Honest assessment of quantum advantage claims	Dequantisation results (Tang 2019, subsequent work): classical algorithms that match quantum speedups for low-rank recommendation problems When quantum advantage survives dequantisation: high-rank, non-sparse data structures where classical sampling fails Benchmarking framework: how to evaluate vendor quantum ML claims against classical baselines on your own data
6	Readiness Assessment and Timeline NISQ limitations and fault-tolerant horizon	Current NISQ hardware constraints: qubit counts, gate fidelities, and coherence times versus production recommendation workloads Fault-tolerant timeline: logical qubit requirements for catalogue-scale quantum ML (estimated 2030-2035 for commercially relevant problem sizes) Near-term value: quantum-inspired classical algorithms (tensor networks, random feature sampling) that deliver today
7	Q&A and Roadmap Discussion

Designed and Delivered By

Workshops are designed and delivered by QSECDEF in collaboration with sector specialists. All facilitators have direct experience in both quantum technologies and digital media systems.

QSECDEF brings world-leading expertise in post-quantum cryptography, quantum computing strategy, and defence-grade security assessment. Our advisory membership spans 600+ organisations and 1,200+ professionals working at the intersection of quantum technologies and critical infrastructure security.

Digital Media workshops are co-delivered with sector specialists who bring direct operational experience in recommendation systems, audience analytics, and ML infrastructure at streaming and publishing scale. This ensures workshop content is grounded in the data volumes, latency requirements, and model serving constraints of production recommendation pipelines.

Commission This Workshop

Sessions are configured around your recommendation architecture, data scale, ML framework, and content catalogue structure. Get in touch to discuss requirements and schedule a date.

Quantum technologies are evolving quickly and new developments emerge regularly. This page was last updated on 15/03/2026. For the most current information about course content and suitability for your organisation, we recommend contacting us directly.

Digital Media Workshops All Consulting Services All Workshops