# Exploring Linear Feature Disentanglement For Neural Networks

arXiv: https://arxiv.org/abs/2203.11700 · HTML: https://mlsanity.com/paper/2203.11700/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 21 references that are in the map.
- First posted on arXiv 22 Mar 2022.

## Builds on (most cited first, 9 of 9)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [Pruning Filters for Efficient ConvNets](https://mlsanity.com/paper/1608.08710/) (2016)
- [Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration](https://mlsanity.com/paper/1811.00250/) (2018)
- [Soft Filter Pruning for Accelerating Deep Convolutional Neural Networks](https://mlsanity.com/paper/1808.06866/) (2018)
- [HRank: Filter Pruning using High-Rank Feature Map](https://mlsanity.com/paper/2002.10179/) (2020)
- [Towards Optimal Structured CNN Pruning via Generative Adversarial Learning](https://mlsanity.com/paper/1903.09291/) (2019)
- [Exploiting Kernel Sparsity and Entropy for Interpretable CNN Compression](https://mlsanity.com/paper/1812.04368/) (2018)
- [Activate or Not: Learning Customized Activation](https://mlsanity.com/paper/2009.04759/) (2020)

## Built on by (most cited first, 3 of 3)

- [Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models](https://mlsanity.com/paper/2403.19647/) (2024)
- [Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks](https://mlsanity.com/paper/2207.13243/) (2022)
- [Are Candidate Models Really Needed for Active Learning?](https://mlsanity.com/paper/2605.14689/) (2026)

## Abstract (arXiv)

Non-linear activation functions, e.g., Sigmoid, ReLU, and Tanh, have achieved great success in neural networks (NNs). Due to the complex non-linear characteristic of samples, the objective of those activation functions is to project samples from their original feature space to a linear separable feature space. This phenomenon ignites our interest in exploring whether all features need to be transformed by all non-linear functions in current typical NNs, i.e., whether there exists a part of features arriving at the linear separable feature space in the intermediate layers, that does not require further non-linear variation but an affine transformation instead. To validate the above hypothesis, we explore the problem of linear feature disentanglement for neural networks in this paper. Specifically, we devise a learnable mask module to distinguish between linear and non-linear features. Through our designed experiments we found that some features reach the linearly separable space earlier than the others and can be detached partly from the NNs. The explored method also provides a readily feasible pruning strategy which barely affects the performance of the original model. We conduct our experiments on four datasets and present promising results.