Trainable Highly-expressive Activation Functions
This addresses the problem of limited expressiveness in neural networks for researchers and practitioners in machine learning, though it is incremental as it builds on existing trainable activation function concepts.
The paper tackles the limitation of fixed activation functions in deep neural networks by introducing DiTAC, a trainable highly-expressive activation function based on an efficient diffeomorphic transformation, which yields substantial improvements in tasks like semantic segmentation, image generation, regression, and image classification.
Nonlinear activation functions are pivotal to the success of deep neural nets, and choosing the appropriate activation function can significantly affect their performance. Most networks use fixed activation functions (e.g., ReLU, GELU, etc.), and this choice might limit their expressiveness. Furthermore, different layers may benefit from diverse activation functions. Consequently, there has been a growing interest in trainable activation functions. In this paper, we introduce DiTAC, a trainable highly-expressive activation function based on an efficient diffeomorphic transformation (called CPAB). Despite introducing only a negligible number of trainable parameters, DiTAC enhances model expressiveness and performance, often yielding substantial improvements. It also outperforms existing activation functions (regardless whether the latter are fixed or trainable) in tasks such as semantic segmentation, image generation, regression problems, and image classification. Our code is available at https://github.com/BGU-CS-VIL/DiTAC.