Rustem Takhanov

activity

20242026

collaborators

Showing stat.MLShow all

2 papers · 1 filter

stat.ML2026

Deep Linear Discriminant Analysis Revisited

Maxat Tezekbayev, Rustem Takhanov, Arman Bolatov +1

We show that for unconstrained Deep Linear Discriminant Analysis (LDA) classifiers, maximum-likelihood training admits pathological solutions in which class means drift together, c…

stat.ML2024

Non-asymptotic spectral bounds on the $ε$ -entropy of kernel classes

Rustem Takhanov

Let $K : I^c ◯ \times I^c ◯$ be a continuous Mercer kernel defined on a compact subset of $R^{n}$ and $H_{K}$ be the reproducing kernel Hilbert space…

Deep Linear Discriminant Analysis Revisited

Non-asymptotic spectral bounds on the ε-entropy of kernel classes

Non-asymptotic spectral bounds on the $ε$ -entropy of kernel classes