MSc Thesis: Dynamic 3D Networks

<b>Overview: </b>A 3D spatially-embedded neural network has parameters conditioned on node coordinates. The network's structure entirely determines its function. The architecture trains via a reinforcement update that transforms a scalar field on R^3. The model was trained on the MNIST image classification dataset, but produced uncompetitive results. Embedding in higher and lower-dimensional spaces is discussed. The author also submitted the project to the ICL Computing department's admissions panel (November 2024).<br><b>File Information:</b> The attached file exhibits an amended version of the author's MSc thesis. The thesis was submitted in completion of the MSc in Artificial Intelligence and Machine Learning at the University of Birmingham in September 2024. The current version was amended for submission to the Imperial College London Computing Department Admission Board in November 2024, and contains purely cosmetic amendments compared to the original.