Johannes Allotey | Searcharxiv

arXiv2026

Extending the SKA Across Africa: The Case for a Continental African VLBI Network

The African continent holds the key to unlocking the full potential of global Very Long Baseline Interferometry (VLBI). Strategic placement of radio telescopes across Africa provides the crucial north-south and intermediate baselines that are currently missing from the global VLBI network. This expansion will dramatically enhance imaging fidelity and resolution. In this chapter, we propose a vision for a continental African VLBI Network (AVN) that will operate in close synergy with SKA-Mid, enabling transformational science across all cosmic scales. While only the Hartebeesthoek Radio Astronomy Observatory (HartRAO) in South Africa and the Ghana Radio Astronomy Observatory (GRAO) are currently operational, several partner countries are in the process of refurbishing or converting existing antennas. Here, we advocate for the expansion of this network through the deployment of a limited number of SKA-Mid-type telescopes across the continent, creating an "African arm" of SKA-VLBI. With a maximum baseline of ~ 9000 km (from Rabat, Morocco to Cassis, Mauritius), the proposed continental facility will surpass for example, the resolution that will be achieved by the next generation Very Large Array (ngVLA) by ~ 10 % at similar observing frequencies and significantly enhance global VLBI coverage. Beyond the scientific and technical gains, the AVN represents a unique opportunity for sustainable growth in human capital, education, and innovation across Africa. Developing and operating a continental VLBI array will train the next generation of engineers, data scientists, and astronomers, stimulate local industry, and inspire public engagement in science and technology. We outline the current status, challenges, and potential roadmap towards realizing this vision, and we highlight how a continental African VLBI network will position Africa at the forefront of global radio astronomy.

astro-ph.IM↗

arXiv2021

Entropy-based Active Learning of Graph Neural Network Surrogate Models for Materials Properties

Graph neural networks, trained on experimental or calculated data are becoming an increasingly important tool in computational materials science. Networks, once trained, are able to make highly accurate predictions at a fraction of the cost of experiments or first-principles calculations of comparable accuracy. However these networks typically rely on large databases of labelled experiments to train the model. In scenarios where data is scarce or expensive to obtain this can be prohibitive. By building a neural network that provides a confidence on the predicted properties, we are able to develop an active learning scheme that can reduce the amount of labelled data required, by identifying the areas of chemical space where the model is most uncertain. We present a scheme for coupling a graph neural network with a Gaussian process to featurise solid-state materials and predict properties \textit{including} a measure of confidence in the prediction. We then demonstrate that this scheme can be used in an active learning context to speed up the training of the model, by selecting the optimal next experiment for obtaining a data label. Our active learning scheme can double the rate at which the performance of the model on a test data set improves with additional data compared to choosing the next sample at random. This type of uncertainty quantification and active learning has the potential to open up new areas of materials science, where data are scarce and expensive to obtain, to the transformative power of graph neural networks.

cond-mat.mtrl-sci↗