On the almost sure convergence of stochastic gradient descent in non-convex problems

P Mertikopoulos, N Hallak, A Kavis… - Advances in Neural …, 2020 - proceedings.neurips.cc
In this paper, we analyze the trajectories of stochastic gradient descent (SGD) with the aim of
understanding their convergence properties in non-convex problems. We first show that the …

Distributed stochastic gradient descent: Nonconvexity, nonsmoothness, and convergence to local minima

B Swenson, R Murray, HV Poor, S Kar - Journal of Machine Learning …, 2022 - jmlr.org
Gradient-descent (GD) based algorithms are an indispensable tool for optimizing modern
machine learning models. The paper considers distributed stochastic GD (D-SGD)--a …

Cb-dsl: Communication-efficient and byzantine-robust distributed swarm learning on non-iid data

X Fan, Y Wang, Y Huo, Z Tian - IEEE Transactions on Cognitive …, 2023 - ieeexplore.ieee.org
The valuable data collected by IoT devices together with the resurgence of machine learning
(ML) stimulate the latest trend of artificial intelligence (AI) at the edge. However, traditional …

Almost sure convergence rates analysis and saddle avoidance of stochastic gradient methods

J Liu, Y Yuan - Journal of Machine Learning Research, 2024 - jmlr.org
The vast majority of convergence rates analysis for stochastic gradient methods in the
literature focus on convergence in expectation, whereas trajectory-wise almost sure …

Online bootstrap inference with nonconvex stochastic gradient descent estimator

Y Zhong, T Kuffner, S Lahiri - ar** saddle points in zeroth-order optimization: the power of two-point estimators
Z Ren, Y Tang, N Li - International Conference on Machine …, 2023 - proceedings.mlr.press
Two-point zeroth order methods are important in many applications of zeroth-order
optimization arising in robotics, wind farms, power systems, online optimization, and …

Computable access control: Embedding access control rules into euclidean space

L Dong, T Wu, W Jia, B Jiang… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
Access control is one of the most basic techniques to ensure the security of the information
system. The traditional access controls of information systems are usually performed based …

3DPG: Distributed deep deterministic policy gradient algorithms for networked multi-agent systems

A Redder, A Ramaswamy, H Karl - arxiv preprint arxiv:2201.00570, 2022 - arxiv.org
We present Distributed Deep Deterministic Policy Gradient (3DPG), a multi-agent actor-critic
(MAAC) algorithm for Markov games. Unlike previous MAAC algorithms, 3DPG is fully …