# Ensemble Mask Networks

arXiv: https://arxiv.org/abs/2309.06382 · HTML: https://mlsanity.com/paper/2309.06382/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 7 references that are in the map.
- First posted on arXiv 12 Sep 2023.

## Builds on (most cited first, 7 of 7)

- [What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimation](https://mlsanity.com/paper/2008.03703/) (2020)
- [Training Neural Networks with Fixed Sparse Masks](https://mlsanity.com/paper/2111.09839/) (2021)
- [Neural Arithmetic Logic Units](https://mlsanity.com/paper/1808.00508/) (2018)
- [An Operator Theoretic View on Pruning Deep Neural Networks](https://mlsanity.com/paper/2110.14856/) (2021)
- [Deep Neural Network Ensembles for Time Series Classification](https://mlsanity.com/paper/1903.06602/) (2019)
- [Neural Graphical Models](https://mlsanity.com/paper/2210.00453/) (2022)
- [A Neural Transfer Function for a Smooth and Differentiable Transition Between Additive and Multiplicative Interactions](https://mlsanity.com/paper/1503.05724/) (2015)

## Abstract (arXiv)

Can an ℝnrightarrow ℝn feedforward network learn matrix-vector multiplication? This study introduces two mechanisms - flexible masking to take matrix inputs, and a unique network pruning to respect the mask's dependency structure. Networks can approximate fixed operations such as matrix-vector multiplication φ(A,x) rightarrow Ax, motivating the mechanisms introduced with applications towards litmus-testing dependencies or interaction order in graph-based models.