Saved in:
Bibliographic Details
Main Authors: Yang, Yi, Lopez, Victor G., Müller, Matthias A.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.20544
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916923094073344
author Yang, Yi
Lopez, Victor G.
Müller, Matthias A.
author_facet Yang, Yi
Lopez, Victor G.
Müller, Matthias A.
contents Beyond the traditional neural network training methods based on gradient descent and its variants, state estimation techniques have been proposed to determine a set of ideal weights from a control-theoretic perspective. Hence, the concept of observability becomes relevant in neural network training. In this paper, we investigate local observability of a class of two-layer feedforward neural networks~(FNNs) with rectified linear unit~(ReLU) activation functions. We analyze local observability of FNNs by evaluating an observability rank condition with respect to the weight matrix and the input sequence. First, we show that, in general, the weights of FNNs are not locally observable. Then, we provide sufficient conditions on the network structures and the weights that lead to local observability. Moreover, we propose an input design approach to render the weights distinguishable and show that this input also excites other weights inside a neighborhood. Finally, we validate our results through a numerical example.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20544
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Local Observability of a Class of Feedforward Neural Networks
Yang, Yi
Lopez, Victor G.
Müller, Matthias A.
Systems and Control
Beyond the traditional neural network training methods based on gradient descent and its variants, state estimation techniques have been proposed to determine a set of ideal weights from a control-theoretic perspective. Hence, the concept of observability becomes relevant in neural network training. In this paper, we investigate local observability of a class of two-layer feedforward neural networks~(FNNs) with rectified linear unit~(ReLU) activation functions. We analyze local observability of FNNs by evaluating an observability rank condition with respect to the weight matrix and the input sequence. First, we show that, in general, the weights of FNNs are not locally observable. Then, we provide sufficient conditions on the network structures and the weights that lead to local observability. Moreover, we propose an input design approach to render the weights distinguishable and show that this input also excites other weights inside a neighborhood. Finally, we validate our results through a numerical example.
title Local Observability of a Class of Feedforward Neural Networks
topic Systems and Control
url https://arxiv.org/abs/2508.20544