Semantic-based Distributed Learning for Diverse and Discriminative Representations

TL;DR

This paper introduces a novel distributed learning framework that produces diverse, discriminative, and structurally meaningful representations, improving collaboration in large-scale, complex data scenarios.

Contribution

The paper proposes a new distributed learning approach that ensures data representations are both diverse and discriminative, with theoretical guarantees and effective clustering strategies.

Findings

The framework maintains discriminative and diverse properties of representations.

Theoretical proof of convergence for i.i.d. data scenarios.

Effective in capturing global structural representations on benchmark datasets.

Abstract

In large-scale distributed scenarios, increasingly complex tasks demand more intelligent collaboration across networks, requiring the joint extraction of structural representations from data samples. However, conventional task-specific approaches often result in nonstructural embeddings, leading to collapsed variability among data samples within the same class, particularly in classification tasks. To address this issue and fully leverage the intrinsic structure of data for downstream applications, we propose a novel distributed learning framework that ensures both diverse and discriminative representations. For independent and identically distributed (i.i.d.) data, we reformulate and decouple the global optimization function by introducing constraints on representation variance. The update rules are then derived and simplified using a primal-dual approach. For non-i.i.d. data distributions, we tackle the problem by clustering and virtually replicating nodes, allowing model updates within each cluster using block coordinate descent. In both cases, the resulting optimal solutions are theoretically proven to maintain discriminative and diverse properties, with a guaranteed convergence for i.i.d. conditions. Additionally, semantic information from representations is shared among nodes, reducing the need for common neural network architectures. Finally, extensive simulations on MNIST, CIFAR-10 and CIFAR-100 confirm the effectiveness of the proposed algorithms in capturing global structural representations.