Bhargava Cube--Inspired Quadratic Regularization for Structured Neural Embeddings
This work addresses the need for interpretable and mathematically consistent embeddings in deep learning, representing an incremental advancement by applying number-theoretic constructs to neural networks.
The paper tackles the problem of unstructured latent spaces in neural representation learning by incorporating algebraic constraints inspired by Bhargava cubes, achieving 99.46% accuracy on MNIST with interpretable 3D embeddings that cluster by digit class.
We present a novel approach to neural representation learning that incorporates algebraic constraints inspired by Bhargava cubes from number theory. Traditional deep learning methods learn representations in unstructured latent spaces lacking interpretability and mathematical consistency. Our framework maps input data to constrained 3-dimensional latent spaces where embeddings are regularized to satisfy learned quadratic relationships derived from Bhargava's combinatorial structures. The architecture employs a differentiable auxiliary loss function operating independently of classification objectives, guiding models toward mathematically structured representations. We evaluate on MNIST, achieving 99.46% accuracy while producing interpretable 3D embeddings that naturally cluster by digit class and satisfy learned quadratic constraints. Unlike existing manifold learning approaches requiring explicit geometric supervision, our method imposes weak algebraic priors through differentiable constraints, ensuring compatibility with standard optimization. This represents the first application of number-theoretic constructs to neural representation learning, establishing a foundation for incorporating structured mathematical priors in neural networks.