The aging of the global population makes the age-friendly design a hot topic in the design field and society. A three-dimensional indoor elderly friendly design system based on adaptive genetic algorithm is proposed t...
详细信息
The rapid advancements in artificial intelligence (AI) have primarily focused on the process of learning from data to acquire knowledge for smart systems. However, the concept of machine unlearning has emerged as a tr...
详细信息
We consider solving a class of unconstrained optimization problems in which only stochastic estimates of the objective functions are available. Existing stochastic optimization methods are mainly extended from gradien...
详细信息
We consider solving a class of unconstrained optimization problems in which only stochastic estimates of the objective functions are available. Existing stochastic optimization methods are mainly extended from gradient-based methods, faced with the challenges of noisy function evaluations, hardness in choosing step-sizes, and probably ill-conditioned landscapes. This paper presents a stochastic evolution strategy (SES) framework and several adaptation schemes to avoid these challenges. The SES framework combines the ideas of population sampling and minibatch sampling in exploiting the zeroth-order gradient information, efficiently reducing the noise in both data selection and gradient approximation. In addition, it admits approximating the gradients using a non-isotropic Gaussian distribution to better capture the curvature information of the landscapes. Based on this framework, we implement a step-size adaptation rule and two covariance matrix adaptation rules, where the former can automatically tune the step-sizes and the latter are intended to cope with ill-conditioning. For SES with certain fixed step-sizes, we establish a nearly optimal convergence rate over smooth landscapes. We also show that using the adaptive step-sizes allows convergence at a slightly slower rate but without the need to know the smoothness constant. Several numerical experiments on machine learning problems verify the above theoretical results and suggest that the adaptive SES methods show much promise.
The Huge Object model for distribution testing, first defined by Goldreich and Ron in 2022, combines the features of classical string testing and distribution testing. In this model we are given access to independent ...
详细信息
Autonomous race driving is a difficult problem, as the driving algorithm needs to find unique maneuvers on the track to achieve the fastest lap time. To date and to the best of our knowledge, there exists no algorithm...
详细信息
In point cloud deep learning, local feature aggregation is crucial to improve the performance of the model. The traditional KNN (K-Nearest Neighbors) method performs local feature aggregation by fixing the K value, bu...
详细信息
Underwater communication is widely regarded as one of the most significant challenges due to the unique physical properties of water. Among the available communication methods, radiofrequency communication offers high...
详细信息
Federated learning has attracted increasing attention with the emergence of distributed data. While extensive federated learning algorithms have been proposed for the non-convex distributed problem, federated learning...
详细信息
ISBN:
(纸本)9781577358800
Federated learning has attracted increasing attention with the emergence of distributed data. While extensive federated learning algorithms have been proposed for the non-convex distributed problem, federated learning in practice still faces numerous challenges, such as the large training iterations to converge since the sizes of models and datasets keep increasing, and the lack of adaptivity by SGD-based model updates. Meanwhile, the study of adaptive methods in federated learning is scarce and existing works either lack a complete theoretical convergence guarantee or have slow sample complexity. In this paper, we propose an efficient adaptive algorithm (i.e., FAFED) based on the momentum-based variancereduced technique in cross-silo FL. We first explore how to design the adaptive algorithm in the FL setting. By providing a counter-example, we prove that a simple combination of FL and adaptive methods could lead to divergence. More importantly, we provide a convergence analysis for our method and prove that our algorithm is the first adaptive FL algorithm to reach the best-known samples O(is an element of(-3)) and O(is an element of(-2)) communication rounds to find an is an element of-stationary point without large batches. The experimental results on the language modeling task and image classification task with heterogeneous data demonstrate the efficiency of our algorithms.
Most of the existing distributed adaptive filtering algorithms over wireless sensor networks (WSNs) are developed, aiming to solve unconstrained network optimization problems. However, in practice, the weight coeffici...
详细信息
Most of the existing distributed adaptive filtering algorithms over wireless sensor networks (WSNs) are developed, aiming to solve unconstrained network optimization problems. However, in practice, the weight coefficients of the filter may need to satisfy a set of linear equations. Thus, a distributed adaptive algorithm that can solve the sensor network optimization problem under constraints is needed. Considering the possible impulsive interference in the observed signals, a novel robust distributed constrained adaptive algorithm called diffusion constrained least mean M-estimate (D-CLMM) is proposed by using the modified Huber function (MHF), which endows the network robustness to impulsive noise. The transient, steady-state performances and stability of the proposed D-CLMM are studied with the aid of some commonly used assumptions and verified by computer simulations. Moreover, the effectiveness of D-CLMM is verified in distributed parameter estimation and beamforming applications in non-Gaussian noise environments.
In this paper, we introduce a versatile data-driven approach for servo-controlling the highly deformable robotic endoscope equipped with Draw Tower Gratings (DTGs). The advancement of perception and decision-making te...
详细信息
暂无评论