# SSDL: Self-Supervised Dictionary Learning

arXiv: https://arxiv.org/abs/2112.01790 · HTML: https://mlsanity.com/paper/2112.01790/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 25 references that are in the map.
- First posted on arXiv 3 Dec 2021.

## Builds on (most cited first, 5 of 5)

- [Label Embedded Dictionary Learning for Image Classification](https://mlsanity.com/paper/1903.03087/) (2019)
- [Sparse, Collaborative, or Nonnegative Representation: Which Helps Pattern Classification?](https://mlsanity.com/paper/1806.04329/) (2018)
- [Hypergraph p-Laplacian Regularization for Remote Sensing Image Recognition](https://mlsanity.com/paper/1806.08104/) (2018)
- [Jointly Learning Structured Analysis Discriminative Dictionary and Analysis Multiclass Classifier](https://mlsanity.com/paper/1905.11543/) (2019)
- [Scalable Block-Diagonal Locality-Constrained Projective Dictionary Learning](https://mlsanity.com/paper/1905.10568/) (2019)

## Abstract (arXiv)

The label-embedded dictionary learning (DL) algorithms generate influential dictionaries by introducing discriminative information. However, there exists a limitation: All the label-embedded DL methods rely on the labels due that this way merely achieves ideal performances in supervised learning. While in semi-supervised and unsupervised learning, it is no longer sufficient to be effective. Inspired by the concept of self-supervised learning (e.g., setting the pretext task to generate a universal model for the downstream task), we propose a Self-Supervised Dictionary Learning (SSDL) framework to address this challenge. Specifically, we first design a p-Laplacian Attention Hypergraph Learning (pAHL) block as the pretext task to generate pseudo soft labels for DL. Then, we adopt the pseudo labels to train a dictionary from a primary label-embedded DL method. We evaluate our SSDL on two human activity recognition datasets. The comparison results with other state-of-the-art methods have demonstrated the efficiency of SSDL.