Workshops Financial Services Financial Fraud Detection

Financial Services Full Day Workshop

Quantum for Financial Fraud Detection

Quantum approaches to fraud detection and AML compliance: anomaly detection algorithms, transaction classification with quantum ML, and graph analysis for anti-money laundering investigations.

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

Covers quantum anomaly detection, quantum machine learning for transaction classification, and quantum graph analysis for AML/KYC compliance. Participants build a quantum fraud detection pipeline, benchmark it against classical baselines, and assess where quantum methods add genuine value in current and near-term fraud analytics workflows.

Financial fraud detection faces a triple challenge: transaction volumes growing faster than classical processing capacity, adversarial actors continuously adapting to detection models, and false positive rates that consume investigation resources. Quantum approaches address different parts of this problem. Quantum anomaly detection algorithms can identify outlier patterns in transaction data using quantum feature spaces that capture correlations classical models miss. Quantum walk algorithms offer polynomial speedups for graph analysis problems central to AML, including entity resolution and beneficial ownership chains. Published proof-of-concept work from the HSBC/IBM quantum fraud detection pilot (2023), the Credit Agricole/Pasqal collaboration on quantum ML for fraud (2024), and academic groups shows measurable improvements in precision on synthetic fraud datasets. Production deployment requires hybrid architectures where quantum handles batch scoring of high-risk segments while classical systems manage real-time throughput.

What participants cover

Quantum anomaly detection: quantum one-class SVM (Rebentrost et al.) and quantum clustering algorithms applied to transaction outlier identification in high-dimensional feature spaces
Variational quantum classifiers for fraud/not-fraud scoring, including quantum kernel methods for feature extraction from transaction metadata and strategies for extreme class imbalance
Hands-on pipeline construction: encoding transaction features, training on labelled fraud data, benchmarking precision and recall against classical random forest and XGBoost
Quantum walk algorithms for AML graph analysis: entity resolution, beneficial ownership chain detection, and network traversal with polynomial speedup over classical methods
Hybrid integration architecture: batch quantum scoring alongside real-time classical systems, with honest assessment of current hardware constraints and maturity level
Published benchmark evidence from the HSBC/IBM pilot and Credit Agricole/Pasqal proof-of-concept, and what these results mean for false positive reduction and production readiness

Preliminary Agenda

Full Day Workshop structure with scheduled breaks. Content is configurable to your organisation's fraud typology, transaction volumes, and existing detection infrastructure.

#	Session	Topics
1	The Fraud Detection Challenge at Scale Transaction volumes, false positive costs, and evolving attack patterns
2	Quantum Anomaly Detection Algorithms Quantum one-class SVM, clustering, and Grover-based pattern search	Quantum one-class SVM for anomaly detection (Rebentrost et al.) and its application to transaction data where classical models plateau against adversarial fraud Quantum clustering algorithms for outlier identification in high-dimensional feature spaces that capture correlations classical models miss Grover-based search for suspicious patterns in large transaction graphs: quadratic speedup, practical circuit depth constraints, and input encoding costs
Break, after 50 min
3	QML for Transaction Classification Variational classifiers and quantum kernel methods for fraud scoring	Variational quantum classifiers for fraud/not-fraud binary classification on transaction data: circuit architecture, parameter training, and convergence behaviour Quantum kernel methods for feature extraction from transaction metadata (amount, frequency, geolocation, merchant category, timestamp patterns) Handling extreme class imbalance: quantum approaches to the 99.9% legitimate transaction problem, including cost-sensitive quantum loss functions
4	Hands-On: Building a Quantum Fraud Detection Pipeline Encoding, training, and benchmarking against classical baselines	Encoding transaction features into quantum states using amplitude and angle encoding strategies Training a variational quantum classifier on labelled fraud data using Qiskit or PennyLane on a simulator backend Measuring precision, recall, and F1 score against classical random forest and XGBoost baselines on the same synthetic dataset
Break, after 60 min
5	AML and KYC: Quantum Approaches to Graph Analysis Quantum walk algorithms for network analysis and entity resolution	Quantum walk algorithms for transaction network analysis: polynomial (not exponential) speedup over classical graph traversal for community detection Entity resolution in transaction graphs: matching identities across fragmented records using quantum similarity measures Beneficial ownership chain detection: tracing layered corporate structures through quantum graph search to identify concealed control relationships
6	Integration Architecture and Honest Assessment Hybrid pipeline design, current benchmarks, and where classical ML remains sufficient	Hybrid classical-quantum pipeline design: quantum batch scoring of high-risk segments alongside real-time classical transaction monitoring Current accuracy benchmarks: proof-of-concept results from the HSBC/IBM pilot (2023) and the Credit Agricole/Pasqal collaboration (2024) on synthetic fraud datasets Honest maturity assessment: false positive reduction is the most likely near-term commercial value; real-time quantum scoring is not feasible on current hardware
7	Q&A and Pilot Planning

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 financial services 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.

Financial Services workshops are co-delivered with sector specialists who bring direct operational experience in financial services organisations. This ensures workshop content is grounded in regulatory, operational, and technical realities specific to the sector.

Commission This Workshop

Sessions are configured around your organisation's fraud typology, transaction volumes, compliance requirements, and existing detection infrastructure. Get in touch to discuss requirements and schedule a date.

Financial Services Workshops All Consulting Services All Workshops