Advances in Distributed Computing and Graph Algorithms

The field of distributed computing and graph algorithms is witnessing significant developments, with a focus on improving the efficiency and scalability of algorithms for various problems. Researchers are exploring new techniques, such as the use of oracles and graph signal processing, to enhance the performance of algorithms for tasks like hierarchical clustering and multi-object tracking. Additionally, there is a growing interest in developing time-aware and synchronization techniques for federated learning and distributed networks. Noteworthy papers include Learning-Augmented Hierarchical Clustering, which presents a polynomial-time algorithm for hierarchical clustering with a splitting oracle, and SyncFed, which introduces a time-aware federated learning framework that employs explicit synchronization and timestamping. Optimizing Cooperative Multi-Object Tracking using Graph Signal Processing is another notable work that proposes a novel framework for tracking objects in 3D LiDAR scenes by formulating and solving a graph topology-aware optimization problem.

Sources

Learning-Augmented Hierarchical Clustering

Perfect Matching with Few Link Activations

Optimal Graph Reconstruction by Counting Connected Components in Induced Subgraphs

Synchronization in Anonymous Networks Under Continuous Dynamics

Pairwise similarity method for majority domination problem

Optimizing Cooperative Multi-Object Tracking using Graph Signal Processing

SyncFed: Time-Aware Federated Learning through Explicit Timestamping and Synchronization

Tight Paths and Tight Pairs in Weighted Directed Graphs

A Unifying Algorithm for Hierarchical Queries

A voice for minorities: diversity in approval-based committee elections under incomplete or inaccurate information

Faster CONGEST Approximation Algorithms for Maximum Weighted Independent Set in Sparse Graphs

Built with on top of