Hardware-Friendly Input Expansion for Accelerating Function Approximation

📅 2026-02-20
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the challenges in one-dimensional function approximation—namely, flat loss landscapes, slow convergence, and difficulty fitting high-frequency components—stemming from parameter symmetry in neural networks. To mitigate these issues, the authors propose a hardware-friendly input space expansion strategy that maps the original one-dimensional input into a higher-dimensional space by incorporating constants such as π. This approach effectively breaks parameter symmetry without increasing the number of model parameters. When combined with the L-BFGS optimizer, the method substantially enhances both training efficiency and approximation accuracy across a range of benchmark one-dimensional functions. Experimental results demonstrate that, with an optimal five-dimensional expansion using π as the expansion constant, the mean squared error is reduced by 66.3% and the average number of L-BFGS iterations decreases by 12%.

Technology Category

Application Category

📝 Abstract
One-dimensional function approximation is a fundamental problem in scientific computing and engineering applications. While neural networks possess powerful universal approximation capabilities, their optimization process is often hindered by flat loss landscapes induced by parameter-space symmetries, leading to slow convergence and poor generalization, particularly for high-frequency components. Inspired by the principle of \emph{symmetry breaking} in physics, this paper proposes a hardware-friendly approach for function approximation through \emph{input-space expansion}. The core idea involves augmenting the original one-dimensional input (e.g., $x$) with constant values (e.g., $\pi$) to form a higher-dimensional vector (e.g., $[\pi, \pi, x, \pi, \pi]$), effectively breaking parameter symmetries without increasing the network's parameter count. We evaluate the method on ten representative one-dimensional functions, including smooth, discontinuous, high-frequency, and non-differentiable functions. Experimental results demonstrate that input-space expansion significantly accelerates training convergence (reducing LBFGS iterations by 12\% on average) and enhances approximation accuracy (reducing final MSE by 66.3\% for the optimal 5D expansion). Ablation studies further reveal the effects of different expansion dimensions and constant selections, with $\pi$ consistently outperforming other constants. Our work proposes a low-cost, efficient, and hardware-friendly technique for algorithm design.
Problem

Research questions and friction points this paper is trying to address.

function approximation
neural networks
symmetry breaking
loss landscape
convergence
Innovation

Methods, ideas, or system contributions that make the work stand out.

input-space expansion
symmetry breaking
function approximation
hardware-friendly
neural networks
🔎 Similar Papers
H
Hu Lou
Northwest Institute of Nuclear Technology, Xi’an, 710000, Shaanxi, China.
Y
Yin-Jun Gao
Northwest Institute of Nuclear Technology, Xi’an, 710000, Shaanxi, China.
D
Dong-Xiao Zhang
Northwest Institute of Nuclear Technology, Xi’an, 710000, Shaanxi, China.
T
Tai-Jiao Du
Northwest Institute of Nuclear Technology, Xi’an, 710000, Shaanxi, China.
J
Jun-Jie Zhang
Northwest Institute of Nuclear Technology, Xi’an, 710000, Shaanxi, China.
J
Jia-Rui Zhang
Northwest Institute of Nuclear Technology, Xi’an, 710000, Shaanxi, China.