Araştırma Makalesi

Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data

Sayı: Advanced Online Publication Erken Görünüm Tarihi: 14 Ağustos 2026
PDF İndir
TR EN

Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data

Öz

ContextMany machine learning and signal processing methods assume Gaussian data because it simplifies analysis and often works reasonably well in practice. But real data are rarely Gaussian. They can have heavy tails, skewness, or unusually large kurtosis, and these differences matter. In fact, non-Gaussianity is exactly what makes problems like Independent Component Analysis (ICA) identifiable in the first place. What’s less clear is how these distributional differences affect learning dynamics, particularly in high-dimensional online settings. Recent work has shown that online ICA can be described using low-dimensional deterministic equations in certain scaling limits. However, most of these analyses assume a fixed source distribution and do not explore what happens when higher-order moments vary systematically. As a result, we still do not fully understand how changes in kurtosis or tail behavior influence stability, convergence speed, or sensitivity to the learning rate.

ObjectiveThis work quantifies how controlled changes in higher-order moment structure shape the macroscopic dynamics of high-dimensional online ICA. We analyze how moment variations interact with initialization and learning rate to determine stability, convergence, and learning speed.

MethodWe analyze a high-dimensional online ICA model through a McKean-type scaling limit, which yields a deterministic measure-valued evolution and an associated closed ordinary differential equation (ODE) for the alignment order parameter. We derive this macroscopic characterization for a broad class of nonlinear contrast functions and specialize the explicit stability and phase-transition analysis to the cubic nonlinearity. We then introduce a moment-controlled source model, constructed as a weighted combination of two non-Gaussian random variables, that preserves zero mean and unit variance while continuously tuning the fourth and sixth moments through a single parameter. In the cubic case, the limiting coefficients depend explicitly on these moments, allowing a detailed phase-plane and stability analysis. We validate the theoretical predictions by comparing the limiting ODE with simulations of the corresponding finite-dimensional online algorithm.

ResultsThe dynamics exhibit moment-induced phase transitions. Larger fourth and sixth moments shrink the basin of attraction of informative solutions, raise initialization thresholds, and reduce the admissible learning-rate range. In the cubic case, the signal-induced drift is governed by excess kurtosis, while the stochastic-damping contribution grows with higher-order moments, producing a trade-off between statistical signal strength and algorithmic stability. Simulations confirm slower convergence, increased sensitivity to initialization, and reduced robustness in large-moment regimes.

ConclusionNon-Gaussianity is necessary for identifiability but can destabilize high-dimensional online learning. Our analysis reveals a fundamental trade-off between statistical richness and algorithmic stability in online ICA. The proposed framework directly links higher-order source statistics to learning dynamics and provides principled insight into learning-rate selection, initialization requirements, and extensions to alternative nonlinearities and multi-source settings.

Anahtar Kelimeler

Proje Numarası

124E063

Etik Beyan

Ethics committee approval is not required for this article.

Kaynakça

  1. S. Åberg, K. Podgórski, “A class of non-Gaussian second order random fields”, Extremes, 14(2), 187–222, 2011. https://doi.org/10.1007/s10687-010-0119-1.
  2. T. W. Lee, M. S. Lewicki, T. J. Sejnowski, “ICA mixture models for unsupervised classification of non-Gaussian classes and automatic context switching in blind signal separation”, IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(10), 1078–1089, 2000. https://doi.org/10.1109/34.879789.
  3. A. Hyvärinen, E. Oja, “Independent component analysis: Algorithms and applications”, Neural Networks, 13(4–5), 411–430, 2000. https://doi.org/10.1016/S0893-6080(00)00026-5.
  4. T. H. Nguyen, U. Şimşekli, M. Gürbüzbalaban, G. Richard, “First Exit Time Analysis of Stochastic Gradient Descent Under Heavy-Tailed Gradient Noise”, Proceedings of the 33rd Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, 8–14 December 2019.
  5. C. J. Li, Z. Wang, H. Liu, “Online ICA: Understanding Global Dynamics of Nonconvex Optimization via Diffusion Processes”, Proceedings of the 30th Conference on Neural Information Processing Systems (NIPS), Barcelona, Spain, 5–11 December 2016, 4967–4975.
  6. Z. Jiao, M. Keller-Ressel, “Emergence of heavy tails in homogenized stochastic gradient descent”, Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, Canada, 10–15 December 2024, 14066–14092. https://doi.org/10.52202/079017-0450.
  7. M. S. Advani, A. M. Saxe, H. Sompolinsky, “High-dimensional dynamics of generalization error in neural networks”, Neural Networks, 132, 428–446, 2020. https://doi.org/10.1016/j.neunet.2020.08.022.
  8. Z. Jin, B. B. Risk, D. S. Matteson, “Optimization and testing in linear non-Gaussian component analysis”, Statistical Analysis and Data Mining, 12(3), 141–156, 2019. https://doi.org/10.1002/sam.11403.

Ayrıntılar

Birincil Dil

İngilizce

Konular

Makine Öğrenmesi Algoritmaları

Bölüm

Araştırma Makalesi

Erken Görünüm Tarihi

14 Ağustos 2026

Yayımlanma Tarihi

-

Gönderilme Tarihi

7 Şubat 2026

Kabul Tarihi

31 Temmuz 2026

Yayımlandığı Sayı

Yıl 2026 Sayı: Advanced Online Publication

Kaynak Göster

APA
Doğan, Z. (2026). Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, Advanced Online Publication. https://doi.org/10.65206/pajes.1884127
AMA
1.Doğan Z. Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi. 2026;(Advanced Online Publication). doi:10.65206/pajes.1884127
Chicago
Doğan, Zafer. 2026. “Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, sy Advanced Online Publication. https://doi.org/10.65206/pajes.1884127.
EndNote
Doğan Z (01 Ağustos 2026) Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi Advanced Online Publication
IEEE
[1]Z. Doğan, “Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data”, Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, sy Advanced Online Publication, Ağu. 2026, doi: 10.65206/pajes.1884127.
ISNAD
Doğan, Zafer. “Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi. Advanced Online Publication (01 Ağustos 2026). https://doi.org/10.65206/pajes.1884127.
JAMA
1.Doğan Z. Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi. 2026. doi:10.65206/pajes.1884127.
MLA
Doğan, Zafer. “Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, sy Advanced Online Publication, Ağustos 2026, doi:10.65206/pajes.1884127.
Vancouver
1.Zafer Doğan. Learning dynamics of high-dimensional online ICA with moment-controlled non-Gaussian Data. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi. 01 Ağustos 2026;(Advanced Online Publication). doi:10.65206/pajes.1884127