2024/10/09 by Ivan Butakov, Butakov, Ivan, Semenenko, Alexander +8 · 2 citations
Computer Science · #94A16 (Primary) 68T07 #94A17 (Secondary) #E.4 #FOS: Computer and information sciences #H.1.1 #Information Theory (cs.IT) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Music and Audio Processing #Speech Recognition and Synthesis #Time Series Analysis and Forecasting
paper · pdf · doi:10.48550/arxiv.2410.06993
openalex publication_date 2024/10/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Deep InfoMax (DIM) is a well-established method for self-supervised representation learning (SSRL) based on maximization of the mutual information between the input and the output of a deep neural network encoder. Despite the DIM and contrastive SSRL in general being well-explored, the task of learning representations conforming to a specific distribution (i.e., distribution matching, DM) is still under-addressed. Motivated by the importance of DM to several downstream tasks (including generative modeling, disentanglement, outliers detection and other), we enhance DIM to enable automatic matching of learned representations to a selected prior distribution. To achieve this, we propose injecting an independent noise into the normalized outputs of the encoder, while keeping the same InfoMax training objective. We show that such modification allows for learning uniformly and normally distributed representations, as well as representations of other absolutely continuous distributions. Our approach is tested on various downstream tasks. The results indicate a moderate trade-off between the performance on the downstream tasks and quality of DM.