The Alan Turing Institute

COMPASS – An open-source software package for creating multi-level representative nested synthetic populations for small areas

Simulation models, such as agent-based or microsimulation models capturing urban contexts, increasingly draw on spatially explicit, attribute-rich synthetic population datasets as real-world data inputs. Despite their growing relevance, the creation of such datasets via simulated...

Bernoulli amputation

A novel, stochastic approach to amputation, the process of introducing missing values to a complete dataset, is presented. It allows one to construct a wide variety of missingness patterns by only having to specify distributions of missingness indicators as opposed to specifying...

Assessing dengue knowledge, attitudes and preventive practices in France and Italy: A cross-sectional study using participatory surveillance cohorts

The dengue virus, transmitted by Aedes mosquitoes, causes an estimated 100–390 million infections annually worldwide. With no treatment available, mitigation and prevention involve reducing mosquito breeding and limiting feeding opportunities. Due to changing climate conditions,...

Cloud Cybersecurity for Research Engineers

The availability of large datasets and breakthroughs in machine learning have had a profound impact on how modern science is conducted. Once the preserve of computer science, large-scale compute has become a universal requirement across research fields from astrophysics to zoolog...

Category-Native Solomonoff Approximation: From Algorithmic Geometry to Kernels, Operators, and Induction

Solomonoff induction mixes all computable explanations with description-length weights, but it is incomputable. This theory-and-position paper argues that practical approximation must be category-native: one should first declare the mathematical category in which a computable sha...

Life in Hut 23: the Turing Research Engineering Group

The Alan Turing Institute's Research Engineering Group (REG) is one of the largest concentrations of research software engineering and data science capability in the UK. This talk is a short tour of how it works in practice and the fascinating range of work we do. If you're in th...

Life in Hut 23: the Turing Research Engineering Group

The Alan Turing Institute's Research Engineering Group (REG) is one of the largest concentrations of research software engineering and data science capability in the UK. This talk is a short tour of how it works in practice and the fascinating range of work we do. If you're in th...

Semi-pessimistic Reinforcement Learning

Offline reinforcement learning aims to learn an optimal policy from pre-collected data. However, it faces challenges of distributional shift, where the learned policy may encounter unseen scenarios not covered in the offline data. Additionally, numerous applications suffer from a...

AutoEmulate: A PyTorch tool for end-to-end emulation workflows

Computational simulations lie at the heart of modern science and engineering, but they are often slow and computationally costly. A common solution is to use emulators: fast, cheap models trained to approximate the simulator. However, constructing these requires substantial exper...

A Self‐Supervised Framework for Space Object Behaviour Characterisation

ABSTRACT Foundation Models, which leverage large neural networks pre‐trained on unlabelled data before fine‐tuning for specific tasks, are increasingly being applied to specialised domains. Recent examples include ClimaX for climate and Clay for satellite Earth observation, but a...

Dual Technology Adoption

Technology adoption remains a critical managerial challenge, in general and specifically in service settings. Investments are substantial but often fail to yield the intended benefits due to low usage among frontline employees, customers, and service partners. Existing technology...