Complex-Valued vs. Real-Valued Neural Networks for Classification Perspectives: An Example on Non-Circular Data

J Agustin A Barrachina (1, 2) , Chengfang Ren (2) , Christele Morisseau (1) , Gilles Vieillard (1) , Jean-Philippe Ovarlez (2, 1)

1 DEMR, ONERA, Université Paris Saclay [Palaiseau]
2 SONDRA - Sondra, CentraleSupélec, Université Paris-Saclay

The contributions of this paper are twofold. First, we show the potential interest of Complex-Valued Neural Network (CVNN) on classification tasks for complex-valued datasets. To highlight this assertion, we investigate an example of complex-valued data in which the real and imaginary parts are statistically dependent through the property of non-circularity. In this context, the performance of fully connected feed-forward CVNNs is compared against a real-valued equivalent model. The results show that CVNN performs better for a wide variety of architectures and data structures. CVNN accuracy presents a statistically higher mean and median and lower variance than Real-Valued Neural Network (RVNN). Furthermore, if no regularization technique is used, CVNN exhibits lower overfitting. The second contribution is the release of a Python library (Barrachina 2019) using Tensorflow as back-end that enables the implementation and training of CVNNs in the hopes of motivating further research on this area.

Domains: Artificial Intelligence [cs.AI]; Machine Learning [cs.LG]; Signal and Image processing; Electronics

J Agustin A Barrachina, Chengfang Ren, Christele Morisseau, Gilles Vieillard, Jean-Philippe Ovarlez. Complex-Valued vs. Real-Valued Neural Networks for Classification Perspectives: An Example on Non-Circular Data. 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Jun 2021, Toronto, Canada. ⟨hal-03144092⟩