A 1/R Law for Kurtosis Contrast in Balanced Mixtures
This addresses a fundamental limitation in ICA for signal processing and machine learning, offering theoretical insights and a practical solution, though it is incremental in refining existing methods.
The paper proves that kurtosis-based ICA weakens in wide, balanced mixtures, showing a sharp redundancy law where excess kurtosis decays as O(1/R), and identifies conditions under which surpassing estimation limits is impossible, while proposing a purification method to restore contrast.
Kurtosis-based Independent Component Analysis (ICA) weakens in wide, balanced mixtures. We prove a sharp redundancy law: for a standardized projection with effective width $R_{\mathrm{eff}}$ (participation ratio), the population excess kurtosis obeys $|κ(y)|=O(κ_{\max}/R_{\mathrm{eff}})$, yielding the order-tight $O(c_bκ_{\max}/R)$ under balance (typically $c_b=O(\log R)$). As an impossibility screen, under standard finite-moment conditions for sample kurtosis estimation, surpassing the $O(1/\sqrt{T})$ estimation scale requires $R\lesssim κ_{\max}\sqrt{T}$. We also show that \emph{purification} -- selecting $m\!\ll\!R$ sign-consistent sources -- restores $R$-independent contrast $Ω(1/m)$, with a simple data-driven heuristic. Synthetic experiments validate the predicted decay, the $\sqrt{T}$ crossover, and contrast recovery.