The present work tackles the money laundering detection problem. A new procedure is introduced which exploits structured time series of both qualitative and quantitative data by means of a transformer neural network. The first step of this procedure aims at learning representations of time series through contrastive learning (without any labels). The second step leverages these representations to generate a money laundering scoring of all observations. A two-thresholds approach is then introduced, which ensures a controlled false-positive rate by means of the Benjamini-Hochberg (BH) procedure. Experiments confirm that the transformer is able to produce general representations that succeed in exploiting money laundering patterns with minimal supervision from domain experts. It also illustrates the higher ability of the new procedure for detecting nonfraudsters as well as fraudsters, while keeping the false positive rate under control. This greatly contrasts with rule-based procedures or the ones based on LSTM architectures.

Regulatory scrutiny of banks’ internal models has intensified, demanding improvements in quantitative risk methodologies and the implementation of new regulatory measures. Repeated portfolio revaluation makes these calculations computationally challenging, notably under the Fundamental Review of the Trading Book (FRTB). In this context, we examine the computational implementation of market risk models, focusing on the practical calculation of Value at Risk (VaR) and Expected Shortfall (ES). Given the complexity of trading portfolios, these models rely on proxies and simplifying assumptions for computational efficiency. This paper explores the use of Gaussian processes (GPs), a powerful machine learning technique, to accelerate the revaluation of derivative portfolios. Trained on datasets generated by sophisticated pricing models, the GP algorithm learns valuation functions and subsequently performs rapid and accurate revaluations. Numerical tests on an equity derivatives portfolio demonstrate that Gaussian process regression (GPR) significantly reduces computation time while maintaining a high accuracy for market risk assessment.