arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2410.05071cs.LGcs.SYeess.SYmath.OCmath.STstat.TH
收藏

带控制应用的随机浅层ReLU网络的函数梯度逼近

Function Gradient Approximation with Random Shallow ReLU Networks with Control Applications

University of Minnesota(明尼苏达大学);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Andrew Lamperski, Siddharth Salapaka

AI总结:

针对控制中浅层ReLU网络函数梯度逼近的理论空白,证明随机输入参数加训练输出参数可实现$O((\text{log}(m)/m)^{1/2})$的梯度误差,改进了常数项并可应用于策略评估问题。

摘要:

神经网络在控制领域被广泛用于逼近未知函数。一种常见的神经网络架构采用单隐藏层(即浅层网络),其中输入参数预先固定,仅训练输出参数。典型的形式化分析表明,若存在输出参数能以足够精度逼近未知函数,则可实现预期的控制性能。一个长期存在的理论空白是,目前尚无条件能保证对于固定的输入参数,通过训练输出参数可达到所需精度。我们近期的工作部分填补了这一空白,证明了若随机选取输入参数,则对于任意足够光滑的函数,以高概率存在输出参数,使得逼近误差为$O((1/m)^{1/2})$,其中$m$为神经元数量。然而,部分应用(尤其是连续时间值函数逼近)要求网络同时以足够精度逼近未知函数及其梯度。本文中,我们证明随机生成的输入参数与经训练的输出参数可带来$O((\text{log}(m)/m)^{1/2})$的梯度误差,此外还改进了我们此前工作中的常数项。我们还展示了如何将该结果应用于策略评估问题。

英文摘要:

Neural networks are widely used to approximate unknown functions in control. A common neural network architecture uses a single hidden layer (i.e. a shallow network), in which the input parameters are fixed in advance and only the output parameters are trained. The typical formal analysis asserts that if output parameters exist to approximate the unknown function with sufficient accuracy, then desired control performance can be achieved. A long-standing theoretical gap was that no conditions existed to guarantee that, for the fixed input parameters, required accuracy could be obtained by training the output parameters. Our recent work has partially closed this gap by demonstrating that if input parameters are chosen randomly, then for any sufficiently smooth function, with high-probability there are output parameters resulting in $O((1/m)^{1/2})$ approximation errors, where $m$ is the number of neurons. However, some applications, notably continuous-time value function approximation, require that the network approximates the both the unknown function and its gradient with sufficient accuracy. In this paper, we show that randomly generated input parameters and trained output parameters result in gradient errors of $O((\log(m)/m)^{1/2})$, and additionally, improve the constants from our prior work. We show how to apply the result to policy evaluation problems.

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。

↑