# Deep Learning of Nonnegativity-Constrained Autoencoders for Enhanced Understanding of Data

arXiv: https://arxiv.org/abs/1802.00003 · HTML: https://mlsanity.com/paper/1802.00003/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jan 2018.

## Built on by (most cited first, 9 of 9)

- [Modeling Content Creator Incentives on Algorithm-Curated Platforms](https://mlsanity.com/paper/2206.13102/) (2022)
- [Fixed points of nonnegative neural networks](https://mlsanity.com/paper/2106.16239/) (2021)
- [The Impact of Activation Sparsity on Overfitting in Convolutional Neural Networks](https://mlsanity.com/paper/2104.06153/) (2021)
- [Multiplicative update rules for accelerating deep learning training and increasing robustness](https://mlsanity.com/paper/2307.07189/) (2023)
- [Fixed Points of Cone Mapping with the Application to Neural Networks](https://mlsanity.com/paper/2207.09947/) (2022)
- [Non-negative isomorphic neural networks for photonic neuromorphic accelerators](https://mlsanity.com/paper/2310.01084/) (2023)
- [Using Part-based Representations for Explainable Deep Reinforcement Learning](https://mlsanity.com/paper/2408.11455/) (2024)
- [Hidden Monotonicity: Explaining Deep Neural Networks via their DC Decomposition](https://mlsanity.com/paper/2601.07700/) (2026)
- [Soft-Constrained Optimization of Latent Space in Variational Autoencoders](https://mlsanity.com/paper/2607.23751/) (2026)

## Abstract (arXiv)

Unsupervised feature extractors are known to perform an efficient and discriminative representation of data. Insight into the mappings they perform and human ability to understand them, however, remain very limited. This is especially prominent when multilayer deep learning architectures are used. This paper demonstrates how to remove these bottlenecks within the architecture of Nonnegativity Constrained Autoencoder (NCSAE). It is shown that by using both L1 and L2 regularization that induce nonnegativity of weights, most of the weights in the network become constrained to be nonnegative thereby resulting into a more understandable structure with minute deterioration in classification accuracy. Also, this proposed approach extracts features that are more sparse and produces additional output layer sparsification. The method is analyzed for accuracy and feature interpretation on the MNIST data, the NORB normalized uniform object data, and the Reuters text categorization dataset.