2025
Leave-One-EquiVariant: Alleviating Invariance-Related Information Loss in Contrastive Music Representations
ICASSP 2025accepted
Contrastive learning has proven effective in self-supervised musical representation learning, particularly for Music Information Retrieval (MIR) tasks. However, reliance on augmentation chains for contrastive view generation and the resulting learnt invariances pose challenges when different downstr…