TeLU Activation Function for Fast and Stable Deep Learning

TL;DR

The paper introduces TeLU, a novel activation function that combines the benefits of ReLU and smoothness, leading to faster, more stable deep learning with improved convergence and robustness.

Contribution

TeLU is a new activation function that approximates the identity in active regions and mitigates vanishing gradients, offering a simple, efficient, and analytically friendly alternative to existing functions.

Findings

TeLU improves convergence speed in deep networks.

TeLU enhances stability and robustness in training.

Experimental results outperform traditional activation functions on benchmarks.

Abstract

We propose the Hyperbolic Tangent Exponential Linear Unit (TeLU), a neural network hidden activation function defined as TeLU(x)=xtanh(exp(x)). TeLU's design is grounded in the core principles of key activation functions, achieving strong convergence by closely approximating the identity function in its active region while effectively mitigating the vanishing gradient problem in its saturating region. Its simple formulation enhances computational efficiency, leading to improvements in scalability and convergence speed. Unlike many modern activation functions, TeLU seamlessly combines the simplicity and effectiveness of ReLU with the smoothness and analytic properties essential for learning stability in deep neural networks. TeLU's ability to mimic the behavior and optimal hyperparameter settings of ReLU, while introducing the benefits of smoothness and curvature, makes it an ideal drop-in replacement. Its analytic nature positions TeLU as a powerful universal approximator, enhancing both robustness and generalization across a multitude of experiments. We rigorously validate these claims through theoretical analysis and experimental validation, demonstrating TeLU's performance across challenging benchmarks; including ResNet18 on ImageNet, Dynamic-Pooling Transformers on Text8, and Recurrent Neural Networks (RNNs) on the Penn TreeBank dataset. These results highlight TeLU's potential to set a new standard in activation functions, driving more efficient and stable learning in deep neural networks, thereby accelerating scientific discoveries across various fields.