# Interpreting Bias in the Neural Networks: A Peek Into Representational Similarity

arXiv: https://arxiv.org/abs/2211.07774 · HTML: https://mlsanity.com/paper/2211.07774/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 24 references that are in the map.
- First posted on arXiv 14 Nov 2022.

## Builds on (most cited first, 16 of 16)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets](https://mlsanity.com/paper/1606.03657/) (2016)
- [Are we done with ImageNet?](https://mlsanity.com/paper/2006.07159/) (2020)
- [Insights on representational similarity in neural networks with canonical correlation](https://mlsanity.com/paper/1806.05759/) (2018)
- [Learning Not to Learn: Training Deep Neural Networks with Biased Data](https://mlsanity.com/paper/1812.10352/) (2018)
- [Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth](https://mlsanity.com/paper/2010.15327/) (2020)
- [REPAIR: Removing Representation Bias by Dataset Resampling](https://mlsanity.com/paper/1904.07911/) (2019)
- [Evaluation of Neural Architectures Trained with Square Loss vs Cross-Entropy in Classification Tasks](https://mlsanity.com/paper/2006.07322/) (2020)
- [Learning Debiased Representation via Disentangled Feature Augmentation](https://mlsanity.com/paper/2107.01372/) (2021)
- [Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biases](https://mlsanity.com/paper/2010.15052/) (2020)
- [Why Do Better Loss Functions Lead to Less Transferable Features?](https://mlsanity.com/paper/2010.16402/) (2020)
- [On Loss Functions for Deep Neural Networks in Classification](https://mlsanity.com/paper/1702.05659/) (2017)
- [Stochastic Bias-Reduced Gradient Methods](https://mlsanity.com/paper/2106.09481/) (2021)
- [On the Versatile Uses of Partial Distance Correlation in Deep Learning](https://mlsanity.com/paper/2207.09684/) (2022)
- [Interpretability of artificial neural network models in artificial Intelligence vs. neuroscience](https://mlsanity.com/paper/2206.03951/) (2022)

## Abstract (arXiv)

Neural networks trained on standard image classification data sets are shown to be less resistant to data set bias. It is necessary to comprehend the behavior objective function that might correspond to superior performance for data with biases. However, there is little research on the selection of the objective function and its representational structure when trained on data set with biases. In this paper, we investigate the performance and internal representational structure of convolution-based neural networks (e.g., ResNets) trained on biased data using various objective functions. We specifically study similarities in representations, using Centered Kernel Alignment (CKA), for different objective functions (probabilistic and margin-based) and offer a comprehensive analysis of the chosen ones. According to our findings, ResNets representations obtained with Negative Log Likelihood (LNLL) and Softmax Cross-Entropy (LSCE) as loss functions are equally capable of producing better performance and fine representations on biased data. We note that without progressive representational similarities among the layers of a neural network, the performance is less likely to be robust.