# Improving performance of recurrent neural network with relu nonlinearity

arXiv: https://arxiv.org/abs/1511.03771 · HTML: https://mlsanity.com/paper/1511.03771/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Nov 2015.

## Built on by (most cited first, 12 of 12)

- [Noisy Recurrent Neural Networks](https://mlsanity.com/paper/2102.04877/) (2021)
- [Tractable Dendritic RNNs for Reconstructing Nonlinear Dynamical Systems](https://mlsanity.com/paper/2207.02542/) (2022)
- [True Zero-Shot Inference of Dynamical Systems Preserving Long-Term Statistics](https://mlsanity.com/paper/2505.13192/) (2025)
- [Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction](https://mlsanity.com/paper/2410.14240/) (2024)
- [Integrating Multimodal Data for Joint Generative Modeling of Complex Dynamics](https://mlsanity.com/paper/2212.07892/) (2022)
- [Optimal Recurrent Network Topologies for Dynamical Systems Reconstruction](https://mlsanity.com/paper/2406.04934/) (2024)
- [NeuroView-RNN: It's About Time](https://mlsanity.com/paper/2202.11811/) (2022)
- [Activity Sparsity Complements Weight Sparsity for Efficient RNN Inference](https://mlsanity.com/paper/2311.07625/) (2023)
- [Evaluation of Time Series Forecasting Models for Estimation of PM2.5 Levels in Air](https://mlsanity.com/paper/2104.03226/) (2021)
- [Gates Are Not What You Need in RNNs](https://mlsanity.com/paper/2108.00527/) (2021)
- [Dynamic Analysis and an Eigen Initializer for Recurrent Neural Networks](https://mlsanity.com/paper/2307.15679/) (2023)
- [Improving the Performance and Learning Stability of Parallelizable RNNs Designed for Ultra-Low Power Applications](https://mlsanity.com/paper/2605.11855/) (2026)

## Abstract (arXiv)

In recent years significant progress has been made in successfully training recurrent neural networks (RNNs) on sequence learning problems involving long range temporal dependencies. The progress has been made on three fronts: (a) Algorithmic improvements involving sophisticated optimization techniques, (b) network design involving complex hidden layer nodes and specialized recurrent layer connections and (c) weight initialization methods. In this paper, we focus on recently proposed weight initialization with identity matrix for the recurrent weights in a RNN. This initialization is specifically proposed for hidden nodes with Rectified Linear Unit (ReLU) non linearity. We offer a simple dynamical systems perspective on weight initialization process, which allows us to propose a modified weight initialization strategy. We show that this initialization technique leads to successfully training RNNs composed of ReLUs. We demonstrate that our proposal produces comparable or better solution for three toy problems involving long range temporal structure: the addition problem, the multiplication problem and the MNIST classification problem using sequence of pixels. In addition, we present results for a benchmark action recognition problem.