Peter Ochieng

2 accepted papers

2025

Diversity Is All You Need for Contrastive Learning: Spectral Bounds on Gradient Magnitudes

NeurIPS 2025poster

Contrastive learning thrives—or fails—based on how we construct \emph{positive} and \emph{negative} pairs. In the absence of explicit labels, models must infer semantic structure from these proxy signals. Early work on Siamese networks \citep{chopra2005learning,hadsell2006dimensionality} already sho…

Cited by 0 Source ↗Scholar ↗

2025

Speech Separation Based on Pre-trained Model and Deep Modularization

Peter Ochieng

ICASSP 2025accepted

This work proposes a speech separation technique that clusters spectrogram points or raw speech blocks by combining traditional graph clustering objectives with deep neural networks. We first extract features from spectrogram points or speech blocks using a pre-trained model, then apply deep modular…

Cited by 0 Source ↗Scholar ↗