P-nets: Deep Polynomial Neural Networks

Polynomial Neural Networks (PNNs) possess a rich algebraic and geometric structure. However, their identifiability-a key property for ensuring interpretabilityremains poorly understood. In this work, we present a comprehensive analysis of the identifiability of deep PNNs, including architectures with and without bias terms. Our results reveal an intricate interplay between activation degrees and layer widths in achieving identifiability. As special cases, we show that architectures with non-increasing layer widths are generically identifiable under mild conditions, while encoder-decoder networks are identifiable when the decoder widths do not grow too rapidly compared to the activation degrees. Our proofs are constructive and center on a connection between deep PNNs and low-rank tensor decompositions, and Kruskal-type uniqueness theorems. We also settle an open conjecture on the dimension of PNN's neurovarieties, and provide new bounds on the activation degrees required for it to reach the expected dimension. * Corresponding author † In this version, the appendices have been reworked for better readability. Appendix E explains the changes between the submitted and the camera-ready version. 39th Conference on Neural Information Processing Systems (NeurIPS 2025).