Non-negative Contrastive Learning
It addresses the problem of opaque features in machine learning for researchers and practitioners, offering an incremental improvement by adapting non-negative matrix factorization principles to contrastive learning.
The paper tackles the lack of interpretability in deep representations by proposing Non-negative Contrastive Learning (NCL), which enforces non-negativity constraints to derive more sparse and disentangled features, resulting in significant outperformance over standard contrastive learning on tasks like feature disentanglement and downstream classification.
Deep representations have shown promising performance when transferred to downstream tasks in a black-box manner. Yet, their inherent lack of interpretability remains a significant challenge, as these features are often opaque to human understanding. In this paper, we propose Non-negative Contrastive Learning (NCL), a renaissance of Non-negative Matrix Factorization (NMF) aimed at deriving interpretable features. The power of NCL lies in its enforcement of non-negativity constraints on features, reminiscent of NMF's capability to extract features that align closely with sample clusters. NCL not only aligns mathematically well with an NMF objective but also preserves NMF's interpretability attributes, resulting in a more sparse and disentangled representation compared to standard contrastive learning (CL). Theoretically, we establish guarantees on the identifiability and downstream generalization of NCL. Empirically, we show that these advantages enable NCL to outperform CL significantly on feature disentanglement, feature selection, as well as downstream classification tasks. At last, we show that NCL can be easily extended to other learning scenarios and benefit supervised learning as well. Code is available at https://github.com/PKU-ML/non_neg.