Nowadays, the variance reduction (VR) technique is used to improve the performance of gradient-type algorithms in machine learning and deep learning. However, some existing VR algorithms require unrealistic assumption...
详细信息
In order to address the convergence issues faced by traditional multi-objective optimization algorithms as the number of objectives increases, this paper proposes a new optimization algorithm, MaOEA-SPC, based on the ...
详细信息
We consider Bayesian optimization using Gaussian Process models, also referred to as kernel-based bandit optimization. We study the methodology of exploring the domain using random samples drawn from a distribution. W...
详细信息
We consider Bayesian optimization using Gaussian Process models, also referred to as kernel-based bandit optimization. We study the methodology of exploring the domain using random samples drawn from a distribution. We show that this random exploration approach achieves the optimal error rates. Our analysis is based on novel concentration bounds in an infinite dimensional Hilbert space established in this work, which may be of independent interest. We further develop an algorithm based on random exploration with domain shrinking and establish its order-optimal regret guarantees under both noise-free and noisy settings. In the noise-free setting, our analysis closes the existing gap in regret performance under a mild assumption on the underlying function and thereby partially resolves a COLT open problem. The proposed algorithm also enjoys a computational advantage over prevailing methods due to the random exploration that obviates the expensive optimization of a non-convex acquisition function for choosing the query points at each iteration. Copyright 2024 by the author(s)
Recently, the use of an unbounded external archive in the design of evolutionary multi-objective optimization (EMO) algorithms has received increasing attention. An important component in the use of an unbounded exter...
详细信息
This paper introduces a methodology for the design of IIR fractional-order digital differentiators with near-linear phase properties. The primary objective is to minimize the maximum phase deviation from the ideal lin...
详细信息
Deep neural networks are powerful and popular learning models;however, recent studies have shown that deep neural network-based policies are susceptible to deception by adversarial attacks. A minimalistic attack is a ...
详细信息
The minimum dominant set problem in graphs is a well-known NP-hard problem that requires the use of heuristics to obtain approximate solutions due to its computational complexity. Among the many heuristic approaches f...
详细信息
In the realm of industrial production, wireless sensor networks (WSNs) have seen extensive application. This paper addresses the significant localization errors associated with traditional DV-Hop algorithms by proposi...
详细信息
In response to the challenges posed by the significant integration of distributed energy resources on distribution network reconfiguration, an improved particle swarm optimization algorithm is proposed in this paper. ...
详细信息
Network optimization and resiliency analysis are pivotal domains revealing network functionality, strength, and resilience. Despite their promise, these methodologies often encounter integration limitations, scalabili...
详细信息
暂无评论