Papp Laszlo, Haberl David, Ecsedi Boglarka, Spielvogel Clemens P, Krajnc Denis, Grahovac Marko, Moradi Sasan, Drexler Wolfgang
Center for Medical Physics and Biomedical Engineering, Medical University of Vienna, Vienna, Austria.
Division of Nuclear Medicine, Medical University of Vienna, Vienna, Austria.
Neural Netw. 2023 Oct;167:517-532. doi: 10.1016/j.neunet.2023.08.026. Epub 2023 Aug 25.
Modern artificial intelligence (AI) approaches mainly rely on neural network (NN) or deep NN methodologies. However, these approaches require large amounts of data to train, given, that the number of their trainable parameters has a polynomial relationship to their neuron counts. This property renders deep NNs challenging to apply in fields operating with small, albeit representative datasets such as healthcare. In this paper, we propose a novel neural network architecture which trains spatial positions of neural soma and axon pairs, where weights are calculated by axon-soma distances of connected neurons. We refer to this method as distance-encoding biomorphic-informational (DEBI) neural network. This concept significantly minimizes the number of trainable parameters compared to conventional neural networks. We demonstrate that DEBI models can yield comparable predictive performance in tabular and imaging datasets, where they require a fraction of trainable parameters compared to conventional NNs, resulting in a highly scalable solution.
现代人工智能（AI）方法主要依赖于神经网络（NN）或深度神经网络方法。然而，这些方法需要大量数据进行训练，因为其可训练参数的数量与其神经元数量呈多项式关系。这一特性使得深度神经网络在诸如医疗保健等使用小数据集（尽管具有代表性）的领域中应用具有挑战性。在本文中，我们提出了一种新颖的神经网络架构，该架构训练神经细胞体和轴突对的空间位置，其中权重由相连神经元的轴突 - 细胞体距离计算得出。我们将此方法称为距离编码生物形态信息（DEBI）神经网络。与传统神经网络相比，这一概念显著减少了可训练参数的数量。我们证明，DEBI模型在表格和成像数据集中可以产生相当的预测性能，与传统神经网络相比，它们所需的可训练参数仅为一小部分，从而产生了一种高度可扩展的解决方案。
