ENN: A Neural Network with DCT Adaptive Activation Functions

TL;DR

ENN introduces DCT-based adaptive activation functions in neural networks, enhancing expressiveness and flexibility, leading to significant performance improvements in classification and regression tasks.

Contribution

The paper proposes a novel DCT-based adaptive activation function model for neural networks, integrating signal processing insights and backpropagation for improved expressiveness.

Findings

Outperforms state-of-the-art benchmarks in accuracy.

Adapts effectively to classification and regression tasks.

Provides high flexibility with low parameter count.

Abstract

The expressiveness of neural networks highly depends on the nature of the activation function, although these are usually assumed predefined and fixed during the training stage. Under a signal processing perspective, in this paper we present Expressive Neural Network (ENN), a novel model in which the non-linear activation functions are modeled using the Discrete Cosine Transform (DCT) and adapted using backpropagation during training. This parametrization keeps the number of trainable parameters low, is appropriate for gradient-based schemes, and adapts to different learning tasks. This is the first non-linear model for activation functions that relies on a signal processing perspective, providing high flexibility and expressiveness to the network. We contribute with insights in the explainability of the network at convergence by recovering the concept of bump, this is, the response of each activation function in the output space. Finally, through exhaustive experiments we show that the model can adapt to classification and regression tasks. The performance of ENN outperforms state of the art benchmarks, providing above a 40% gap in accuracy in some scenarios.