Songbai Liu

Population-Aware Contrastive Surrogates for Expensive Multiobjective Optimization

Songbai Liu, Xun Zhou, Hao Sun, Lijia Ma et autres

cn (code pays fourni par la source)

Progressive Colour Equalisation and Detail Refinement for Underwater Image Enhancement

Songbai Liu, Jiachen Huang

ABSTRACT Underwater image enhancement remains a critical challenge in computational vision due to complex distortions caused by wavelength‐dependent light absorption and scattering. This paper introduces CEDFNet, a novel two‐stage framework that leverages advanced computational intelligence techniques for robust and high‐fidelity underwater image …

cn (code pays fourni par la source)

FedBPG: Balancing Generalization and Personalization in Federated Learning for Consumer Electronics

Songbai Liu, Heping Liu, Lijia Ma, Qiuzhen Lin et autres

In the realm of consumer electronics, data surges have caused critical data privacy and integrity issues. Federated learning (FL) emerges as a promising solution, allowing devices to collaboratively train models without transmitting raw data, thus preserving privacy. However, data heterogeneity in practical …

cn (code pays fourni par la source)

Multiobjective Many-Tasking Evolutionary Optimization Based on Optimizer Portfolio

Wu Lin, Weitao Chen, Qiaoling Zhan, Songbai Liu et autres

cn, hk (code pays fourni par la source)

Heterogeneous Federated Learning Using Reliable Prototypes with Graph Neural Networks

Qingling Zhu, Shaifeng Zheng, Songbai Liu, Junkai Ji et autres

cn, hk (code pays fourni par la source)

Topology-Weighted Effective Rank: A Zero-Cost Proxy for Training Dynamics Stability in Neural Architecture Search

Haojie Zhang, Yeming Yang, Songbai Liu, Lijia Ma et autres

cn, hk (code pays fourni par la source)

Multipattern Learning and Collaboration-Based Evolutionary Optimizer for Large-Scale Multiobjective Optimization

Wei Song, Mingshuo Song, Haojie Zhou, Xiaoyan Sun et autres

Recently, machine learning-embedded large-scale multiobjective evolutionary algorithms (LMOEAs) have shown great promise in solving large-scale multiobjective optimization problems (LMOPs). However, the fast convergence of the population to the true Pareto-optimal front (POF) and even distribution of the obtained Pareto-optimal solutions (POSs) on …

cn, gb (code pays fourni par la source)

Multi-Population Evolutionary Neural Architecture Search via Multiple Zero-Cost Proxies

Jiehui Feng, Wu Lin, Qingling Zhu, Songbai Liu et autres

Zero-cost proxies have attracted growing attention in neural architecture search (NAS) for their efficiency in evaluating neural architectures at low computational cost. However, empirical studies have revealed that zero-cost proxies exhibit inherent biases, leading to performance degradation in complex scenarios. To alleviate …

cn (code pays fourni par la source)

Evolutionary Transfer Optimization Assisted by Unselected Features for Multiobjective Feature Selection

Songbai Liu, Xuan Duan, Lijia Ma, Qiuzhen Lin et autres

Feature selection plays a crucial role in classification tasks, particularly in high-dimensional datasets where identifying relevant features while minimizing redundancy is challenging. Traditional multiobjective feature selection (MOFS) methods face challenges due to random initialization and focusing solely on the classification accuracy of …

cn, hk (code pays fourni par la source)

An Enhanced Search Direction-Based Knowledge Transfer for Multiobjective Many-Tasking Evolutionary Optimization

Lin Wu, Songbai Liu, Qingling Zhu, Qiuzhen Lin

This paper proposes a multiobjective many-tasking evolutionary algorithm with enhanced search direction-based knowledge transfer (MMaTEA-ESD). Specifically, the search directions for all tasks are first dynamically computed based on the populations obtained during the evolutionary search process. Subsequently, the search direction of the …

cn (code pays fourni par la source)

Multifaceted collaborative evolutionary constrained multimodal multiobjective optimization

Zeyi Wang, Songbai Liu, Lijia Ma, Qiuzhen Lin et autres

cn (code pays fourni par la source)

Clustering-Based Evolutionary Federated Multiobjective Optimization and Learning

Songbai Liu

Federated learning enables decentralized model training while preserving data privacy, yet it faces challenges in balancing communication efficiency, model performance, and privacy protection. To address these trade-offs, we formulate FL as a federated multiobjective optimization problem and propose FedMOEAC, a clustering-based evolutionary …