← Search

Octavian-Vlad Murad

2 accepted papers

2025

The Noisy Laplacian: a Threshold Phenomenon for Non-Linear Dimension Reduction

ICML 2025poster

In this paper, we clarify the effect of noise on common spectrally motivated algorithms such as Diffusion Maps (DM) for dimension reduction. Empirically, these methods are much more robust to noise than current work suggests. Specifically, existing consistency results require that either the noise a…

Cited by 0SourcePDFScholar
2024

Consistency of Dictionary-Based Manifold Learning

AISTATS 2024poster

We analyze a paradigm for interpretable Manifold Learning for scientific data analysis, whereby one parametrizes a manifold with d smooth functions from a scientist-provided dictionary of meaningful, domain-related functions. When such a parametrization exists, we provide an algorithm for finding it…

Cited by 2SourcePDFScholar