Classical and Quantum Speedups for Non-Convex Optimization via Energy Conserving Descent

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sun, Yihang, Wang, Huaijin, Hayden, Patrick, Blanchet, Jose
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914472638021632
author Sun, Yihang
Wang, Huaijin
Hayden, Patrick
Blanchet, Jose
author_facet Sun, Yihang
Wang, Huaijin
Hayden, Patrick
Blanchet, Jose
contents The Energy Conserving Descent (ECD) algorithm was recently proposed (De Luca & Silverstein, 2022) as a global non-convex optimization method. Unlike gradient descent, appropriately configured ECD dynamics escape strict local minima and converge to a global minimum, making it appealing for machine learning optimization. We present the first analytical study of ECD, focusing on the one-dimensional setting for this first installment. We formalize a stochastic ECD dynamics (sECD) with energy-preserving noise, as well as a quantum analog of the ECD Hamiltonian (qECD), providing the foundation for a quantum algorithm through Hamiltonian simulation. For positive double-well objectives, we compute the expected hitting time from a local to the global minimum. We prove that both sECD and qECD yield exponential speedup over respective gradient descent baselines--stochastic gradient descent and its quantization. For objectives with tall barriers, qECD achieves a further speedup over sECD.
format Preprint
id arxiv_https___arxiv_org_abs_2604_13022
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Classical and Quantum Speedups for Non-Convex Optimization via Energy Conserving Descent
Sun, Yihang
Wang, Huaijin
Hayden, Patrick
Blanchet, Jose
Quantum Physics
Machine Learning
Optimization and Control
The Energy Conserving Descent (ECD) algorithm was recently proposed (De Luca & Silverstein, 2022) as a global non-convex optimization method. Unlike gradient descent, appropriately configured ECD dynamics escape strict local minima and converge to a global minimum, making it appealing for machine learning optimization. We present the first analytical study of ECD, focusing on the one-dimensional setting for this first installment. We formalize a stochastic ECD dynamics (sECD) with energy-preserving noise, as well as a quantum analog of the ECD Hamiltonian (qECD), providing the foundation for a quantum algorithm through Hamiltonian simulation. For positive double-well objectives, we compute the expected hitting time from a local to the global minimum. We prove that both sECD and qECD yield exponential speedup over respective gradient descent baselines--stochastic gradient descent and its quantization. For objectives with tall barriers, qECD achieves a further speedup over sECD.
title Classical and Quantum Speedups for Non-Convex Optimization via Energy Conserving Descent
topic Quantum Physics
Machine Learning
Optimization and Control
url https://arxiv.org/abs/2604.13022