Kuan-Ching Li

IKCS: An Effective Intelligent K-Medoids ClusteringScheme Based on Granularity Rough Entropy

k-Medoids Clustering algorithm is one of the most commonly used clustering methods. However, it does not consider the weight of each feature, and the selection of center points is random, which may affect the clustering results. Granularity computing analyzes knowledge from various perspectives, considering the level o...

Segment-Account Migration: Efficient Account Migration Based on Lock-Free Mechanism

Blockchain represents a critical technical solution for achieving reliable and trustworthy distributed storage in JointCloud computing. Sharding technology can enhance the scalability of blockchain. However, while improving throughput, the massive cross-shard transactions introduced by sharding have become a new perfor...

Communication-Efficient Federated Learning Based on a Two-Stage Adaptive Basis Matrix for Gradient Compression

Federated learning (FL) faces significant communication overhead due to the repeated exchange of large gradient tensors between clients and the server. While existing compression techniques, such as sparsification and quantization, help reduce this cost, they often result in information loss, which can negatively impac...

Lightweight and robust audio-visual emotion recognition via multi-scale mamba temporal modeling and quality-aware expert fusion

Experiments demonstrate that the proposed lightweight audio-visual emotion recognition framework achieves competitive recognition accuracy with significantly fewer parameters and lower computational cost.

MultiSecDFL: Multi-Metric Filtering-Based Secure Aggregation Method for Decentralized Federated Learning