# On the Selection of Initialization and Activation Function for Deep Neural Networks

arXiv: https://arxiv.org/abs/1805.08266 · HTML: https://mlsanity.com/paper/1805.08266/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 May 2018.

## Built on by (most cited first, 11 of 11)

- [Model Complexity of Deep Learning: A Survey](https://mlsanity.com/paper/2103.05127/) (2021)
- [KANQAS: Kolmogorov-Arnold Network for Quantum Architecture Search](https://mlsanity.com/paper/2406.17630/) (2024)
- [Phase diagram of early training dynamics in deep neural networks: effect of the learning rate, depth, and width](https://mlsanity.com/paper/2302.12250/) (2023)
- [A Deep Collocation Method for the Bending Analysis of Kirchhoff Plate](https://mlsanity.com/paper/2102.02617/) (2021)
- [Understanding the Expressivity and Trainability of Fourier Neural Operator: A Mean-Field Perspective](https://mlsanity.com/paper/2310.06379/) (2023)
- [TranSHER: Translating Knowledge Graph Embedding with Hyper-Ellipsoidal Restriction](https://mlsanity.com/paper/2204.13221/) (2022)
- [Frequency Disentangled Residual Network](https://mlsanity.com/paper/2109.12556/) (2021)
- [Criticality versus uniformity in deep neural networks](https://mlsanity.com/paper/2304.04784/) (2023)
- [Initializing ReLU networks in an expressive subspace of weights](https://mlsanity.com/paper/2103.12499/) (2021)
- [Impact of Tuning Parameters in Deep Convolutional Neural Network Using a Crack Image Dataset](https://mlsanity.com/paper/2506.03184/) (2025)
- [Batch Matrix-form Equations and Implementation of Multilayer Perceptrons](https://mlsanity.com/paper/2511.11918/) (2025)

## Abstract (arXiv)

The weight initialization and the activation function of deep neural networks have a crucial impact on the performance of the training procedure. An inappropriate selection can lead to the loss of information of the input during forward propagation and the exponential vanishing/exploding of gradients during back-propagation. Understanding the theoretical properties of untrained random networks is key to identifying which deep networks may be trained successfully as recently demonstrated by Schoenholz et al. (2017) who showed that for deep feedforward neural networks only a specific choice of hyperparameters known as the `edge of chaos' can lead to good performance. We complete this analysis by providing quantitative results showing that, for a class of ReLU-like activation functions, the information propagates indeed deeper for an initialization at the edge of chaos. By further extending this analysis, we identify a class of activation functions that improve the information propagation over ReLU-like functions. This class includes the Swish activation, φswish(x) = x · sigmoid(x), used in Hendrycks & Gimpel (2016), Elfwing et al. (2017) and Ramachandran et al. (2017). This provides a theoretical grounding for the excellent empirical performance of φswish observed in these contributions. We complement those previous results by illustrating the benefit of using a random initialization on the edge of chaos in this context.