Chisme: Fully Decentralized Differentiated Deep Learning for Edge Intelligence

2025-05-14

AI Summary

To address resource constraints, intermittent connectivity, data heterogeneity, and infrastructure scarcity in edge networks, this paper proposes Chisme—a dual-paradigm decentralized deep learning framework supporting collaborative personalized learning: synchronous Chisme-DFL and asynchronous Chisme-GL. Its core innovation is the first-ever heuristic for data affinity based on model update similarity, enabling dynamic, adaptive weighted aggregation. Additionally, it introduces the first co-designed protocol pair: a linearly scalable synchronous protocol and a constant-overhead asynchronous protocol. Evaluated across diverse network topologies—including sparse, unreliable, and fully connected graphs—Chisme outperforms standard decentralized federated learning (DFL) and gossip-based learning (GL) in model accuracy and personalization, achieving a 12.7% improvement in personalized performance while reducing communication and memory overhead by 31% and 24%, respectively.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Distributed Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Decentralized Web and Fediverse systems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

As demand for intelligent services rises and edge devices become more capable, distributed learning at the network edge has emerged as a key enabling technology. While existing paradigms like federated learning (FL) and decentralized FL (DFL) enable privacy-preserving distributed learning in many scenarios, they face potential challenges in connectivity and synchronization imposed by resource-constrained and infrastructure-less environments. While more robust, gossip learning (GL) algorithms have generally been designed for homogeneous data distributions and may not suit all contexts. This paper introduces Chisme, a novel suite of protocols designed to address the challenges of implementing robust intelligence in the network edge, characterized by heterogeneous data distributions, episodic connectivity, and lack of infrastructure. Chisme includes both synchronous DFL (Chisme-DFL) and asynchronous GL (Chisme-GL) variants that enable collaborative yet decentralized model training that considers underlying data heterogeneity. We introduce a data similarity heuristic that allows agents to opportunistically infer affinity with each other using the existing communication of model updates in decentralized FL and GL. We leverage the heuristic to extend DFL's model aggregation and GL's model merge mechanisms for better personalized training while maintaining collaboration. While Chisme-DFL is a synchronous decentralized approach whose resource utilization scales linearly with network size, Chisme-GL is fully asynchronous and has a lower, constant resource requirement independent of network size. We demonstrate that Chisme methods outperform their standard counterparts in model training over distributed and heterogeneous data in network scenarios ranging from less connected and reliable networks to fully connected and lossless networks.

Problem

• Addresses decentralized learning challenges in edge networks
• Handles heterogeneous data distributions and episodic connectivity
• Improves model training for resource-constrained environments

Innovation

• Decentralized differentiated learning for edge intelligence
• Data similarity heuristic for personalized training
• Synchronous and asynchronous variants for scalability