Enregistré dans:
| Auteur principal: | |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2605.08171 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909028132585472 |
|---|---|
| author | Pan, Lurong |
| author_facet | Pan, Lurong |
| contents | Background and motivation. The Communication Dynamics (CD) framework, introduced in two earlier papers for atomic-energy prediction and field-induced superconductivity, treats each physical channel as a (2l+1)-vertex polygon whose discrete Fourier transform yields its energy spectrum. This paper applies the same circulant-spectral machinery to neural-network design.
Layer construction. CDLinear is a block-circulant linear layer with block size B = 2l+1 and 1/B the parameter count of a dense layer of equal input/output dimensions. Three properties follow from the construction. (i) The Hessian of mean-squared loss with respect to the weights is diagonalized by the discrete Fourier transform, with eigenvalues |F[Xj](k)|^2 read directly from the input statistics (Theorem 1). (ii) Under input pre-whitening, the population Hessian condition number satisfies kappa = 1 exactly, with the empirical condition number bounded by 1+O(sqrt(B/N)) on N samples (Theorem 2). (iii) The Shannon noise rate alpha_CD = 0.0118 calibrated in the parent CD papers from the Na D-doublet specifies a transferable, non-arbitrary dropout rate.
Empirical evaluation. A CDLinear MLP at B = 4 achieves 97.50% +/- 0.23% test accuracy with 2,380 parameters versus 98.15% +/- 0.47% for a parameter-matched dense MLP at 8,970 parameters, a 3.8x parameter reduction at 0.65% accuracy cost, within one standard deviation of the seed-to-seed spread. The CD-MLP mean Hessian condition number kappa = 1.9x10^4 is 310x smaller than the dense baseline kappa = 5.9x10^6, in quantitative agreement with Theorem 2. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_08171 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Communication Dynamics Neural Networks: FFT-Diagonalized Layers for Improved Hessian Conditioning at Reduced Parameter Count Pan, Lurong Machine Learning Artificial Intelligence Background and motivation. The Communication Dynamics (CD) framework, introduced in two earlier papers for atomic-energy prediction and field-induced superconductivity, treats each physical channel as a (2l+1)-vertex polygon whose discrete Fourier transform yields its energy spectrum. This paper applies the same circulant-spectral machinery to neural-network design. Layer construction. CDLinear is a block-circulant linear layer with block size B = 2l+1 and 1/B the parameter count of a dense layer of equal input/output dimensions. Three properties follow from the construction. (i) The Hessian of mean-squared loss with respect to the weights is diagonalized by the discrete Fourier transform, with eigenvalues |F[Xj](k)|^2 read directly from the input statistics (Theorem 1). (ii) Under input pre-whitening, the population Hessian condition number satisfies kappa = 1 exactly, with the empirical condition number bounded by 1+O(sqrt(B/N)) on N samples (Theorem 2). (iii) The Shannon noise rate alpha_CD = 0.0118 calibrated in the parent CD papers from the Na D-doublet specifies a transferable, non-arbitrary dropout rate. Empirical evaluation. A CDLinear MLP at B = 4 achieves 97.50% +/- 0.23% test accuracy with 2,380 parameters versus 98.15% +/- 0.47% for a parameter-matched dense MLP at 8,970 parameters, a 3.8x parameter reduction at 0.65% accuracy cost, within one standard deviation of the seed-to-seed spread. The CD-MLP mean Hessian condition number kappa = 1.9x10^4 is 310x smaller than the dense baseline kappa = 5.9x10^6, in quantitative agreement with Theorem 2. |
| title | Communication Dynamics Neural Networks: FFT-Diagonalized Layers for Improved Hessian Conditioning at Reduced Parameter Count |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2605.08171 |