# Poster: Link between Bias, Node Sensitivity and Long-Tail Distribution in trained DNNs

arXiv: https://arxiv.org/abs/2303.16589 · HTML: https://mlsanity.com/paper/2303.16589/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 23 references that are in the map.
- First posted on arXiv 29 Mar 2023.

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

- [Deep Representation Learning on Long-tailed Data: A Learnable Embedding Augmentation Perspective](https://mlsanity.com/paper/2002.10826/) (2020)
- [Distributional Robustness Loss for Long-tail Learning](https://mlsanity.com/paper/2104.03066/) (2021)
- [Fairness Through Robustness: Investigating Robustness Disparity in Deep Learning](https://mlsanity.com/paper/2006.12621/) (2020)
- [Hate Speech Detection: A Solved Problem? The Challenging Case of Long Tail on Twitter](https://mlsanity.com/paper/1803.03662/) (2018)
- [UnbiasedNets: A Dataset Diversification Framework for Robustness Bias Alleviation in Neural Networks](https://mlsanity.com/paper/2302.12538/) (2023)

## Abstract (arXiv)

Owing to their remarkable learning (and relearning) capabilities, deep neural networks (DNNs) find use in numerous real-world applications. However, the learning of these data-driven machine learning models is generally as good as the data available to them for training. Hence, training datasets with long-tail distribution pose a challenge for DNNs, since the DNNs trained on them may provide a varying degree of classification performance across different output classes. While the overall bias of such networks is already highlighted in existing works, this work identifies the node bias that leads to a varying sensitivity of the nodes for different output classes. To the best of our knowledge, this is the first work highlighting this unique challenge in DNNs, discussing its probable causes, and providing open challenges for this new research direction. We support our reasoning using an empirical case study of the networks trained on a real-world dataset.