arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.17142math.OCcs.LGcs.SYeess.SY
收藏

随机浅层ReLU网络的逼近及其在模型参考自适应控制中的应用

Approximation with Random Shallow ReLU Networks with Applications to Model Reference Adaptive Control

University of Minnesota(明尼苏达大学);Honeywell Aerospace(霍尼韦尔航空航天公司)

Andrew Lamperski, Tyler Lekang

AI总结:

该研究证明随机权重偏置的浅层ReLU网络对光滑函数有高概率的$O(m^{-1/2})$的$L_{\nfty}$逼近误差,并将其应用于模型参考自适应控制以获取指定精度的逼近,填补了相关理论空白。

摘要:

神经网络常被用于非线性系统的自适应控制及相关强化学习方法中。一种常见架构采用单隐藏层神经网络(即浅层网络),其权重和偏置预先固定,仅输出层接受训练。经典结果虽表明存在此类神经网络可在有界区域上逼近任意连续函数,但这些结果是非构造性的,且实际使用的网络并无逼近保证。因此,神经网络控制所需的逼近性质是被假设的,而非得到证明的。本文旨在填补这一空白,证明对于足够光滑的函数,权重和偏置随机生成的ReLU网络能以高概率达到$O(m^{-1/2})$的$L_{\infty}$误差,其中$m$为神经元数量。只需在球面上均匀生成权重、在区间上均匀生成偏置即可满足要求。本文还展示了该结果如何用于在模型参考自适应控制应用中获得所需精度的逼近。

英文摘要:

Neural networks are regularly employed in adaptive control of nonlinear systems and related methods of reinforcement learning. A common architecture uses a neural network with a single hidden layer (i.e. a shallow network), in which the weights and biases are fixed in advance and only the output layer is trained. While classical results show that there exist neural networks of this type that can approximate arbitrary continuous functions over bounded regions, they are non-constructive, and the networks used in practice have no approximation guarantees. Thus, the approximation properties required for control with neural networks are assumed, rather than proved. In this paper, we aim to fill this gap by showing that for sufficiently smooth functions, ReLU networks with randomly generated weights and biases achieve $L_{\infty}$ error of $O(m^{-1/2})$ with high probability, where $m$ is the number of neurons. It suffices to generate the weights uniformly over a sphere and the biases uniformly over an interval. We show how the result can be used to get approximations of required accuracy in a model reference adaptive control application.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑