• 15:00 – 16:00
Dr Alessandro Micheli, Imperial College London
Title: NeuralSurv: Deep Survival Analysis with Bayesian Uncertainty Quantification
We introduce NeuralSurv, the first deep survival model to incorporate Bayesian uncertainty quantification. Our non parametric, architecture agnostic framework captures time varying covariate–risk relationships in continuous time via a novel two stage data augmentation scheme, for which we establish theoretical guarantees. For efficient posterior inference, we introduce a mean field variational algorithm with coordinate ascent updates that scale linearly in model size. By locally linearizing the Bayesian neural network, we obtain full conjugacy and derive all coordinate updates in closed form.
In experiments, NeuralSurv delivers superior calibration compared to state-of-the-art deep survival models, while matching or exceeding their discriminative performance across both synthetic benchmarks and real-world datasets.Our results demonstrate the value of Bayesian principles in data scarce regimes by enhancing model calibration and providing robust, well calibrated uncertainty estimates for the survival function.
16:05 – 16:50
Robert Boyce, Imperial College London
Title: Market making - internalisation and competition
The problem of liquidity provision is of fundamental interest for the operation of financial markets and clients’ transaction costs. Dealers provide liquidity and in doing so take on so-called ‘inventory risk’ in return for ‘making the spread’. In this talk, we review the problem of liquidity provision from a mathematical finance perspective and consider extensions incorporating features of real-world markets such as competition between dealers, the internalisation/externalisation trade-off, and the emergence of internal exchanges in the FX market.
16:55 -17:40
Nikkita Ngalande, Oxford University
Title: Network-Based Analysis of Systemic Risk in Economic Networks
Economic shocks propagate through supply-chain networks via inter-sectoral dependencies, but classical Leontief analysis captures only demand-side transmission. We develop a dual-propagation framework integrating both Leontief (demand) and Ghosh (supply) inverses on the same network, revealing fundamentally different spectral properties and propagation patterns. This provides mathematical foundations for quantifying systemic risk in interconnected economic systems.
