analysis-gradient-descent-circuits
Code & Simulink models for “Analysis of gradient descent algorithms: Discrete to continuous domains and circuit equivalents” (SCL, 2025)
https://github.com/HAOHE123/analysis-gradient-descent-circuits
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (3.4%) to scientific vocabulary
Keywords
Repository
Code & Simulink models for “Analysis of gradient descent algorithms: Discrete to continuous domains and circuit equivalents” (SCL, 2025)
Basic Info
- Host: GitHub
- Owner: HAOHE123
- License: mit
- Language: Python
- Default Branch: main
- Homepage: https://doi.org/10.1016/j.sysconle.2025.106146
- Size: 170 KB
Statistics
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
- Releases: 0
Topics
Metadata Files
README.md
Analysis of Gradient Descent Algorithms: Discrete ↔ Continuous Domains & Circuit Equivalents
Code and examples for “Analysis of gradient descent algorithms: Discrete to continuous domains and circuit equivalents”
Overview
This repository contains gradient functions used for the paper:
He Hao, Daniel Silvestre, Carlos Silvestre, “Analysis of gradient descent algorithms: Discrete to continuous domains and circuit equivalents,” Systems & Control Letters, June 4, 2025.
The code covers:
- Discrete‐time gradient descent methods.
- Continuous‐time ODE counterparts for these algorithms.
- Circuit‐equivalent Simulink models
Owner
- Name: He Hao
- Login: HAOHE123
- Kind: user
- Location: Lisboa, Portugal
- Company: Instituto Superior Técnico
- Repositories: 1
- Profile: https://github.com/HAOHE123
Ph.D. Student at Instituto Superior Técnico in Electrical and Computer Engineering
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you use this software, please cite:"
title: "Analysis of gradient descent algorithms: Discrete to continuous domains and circuit equivalents"
version: "v1.0"
doi: "10.1016/j.sysconle.2025.106146"
url: "https://github.com/HAOHE123/Continuous-Gradient-Descent-algorithms"
authors:
- family-names: "Hao"
given-names: "He"
- family-names: "Silvestre"
given-names: "Daniel"
- family-names: "Silvestre"
given-names: "Carlos"
date-released: "2025-06-04"
license: "MIT"
keywords:
- "Optimization-based controllers"
- "Iterative solvers"
abstract: |
In recent years, there have been several advances in iterative optimization algorithms seen as closed-loop control systems in discrete-time that solve unconstrained optimization problems. In this paper, we extend these advances to the continuous setting and leverage circuit equivalence to present a possible implementation of such controllers. Next, we address constrained Quadratic Programming (QP) challenges within a primal–dual framework that appears in many controller definitions. By drawing parallels between second-order ODEs and circuit dynamics, our study bridges theoretical optimization with practical electrical analogues that can be designed and implemented for problems in systems engineering, robotics, and autonomous vehicles that could benefit from the low power and latency of these optimization-based controllers.