Cross-Gradient Aggregation for Decentralized Learning from Non-IID data
This addresses the challenge of decentralized learning for collaborative agents in real-world applications where data is non-IID, offering an incremental improvement over existing algorithms.
The paper tackles the problem of decentralized learning with non-IID data by proposing Cross-Gradient Aggregation (CGA), which aggregates cross-gradient information and uses quadratic programming for updates, achieving superior performance over state-of-the-art methods on MNIST and CIFAR-10 datasets while maintaining efficiency under communication compression.
Decentralized learning enables a group of collaborative agents to learn models using a distributed dataset without the need for a central parameter server. Recently, decentralized learning algorithms have demonstrated state-of-the-art results on benchmark data sets, comparable with centralized algorithms. However, the key assumption to achieve competitive performance is that the data is independently and identically distributed (IID) among the agents which, in real-life applications, is often not applicable. Inspired by ideas from continual learning, we propose Cross-Gradient Aggregation (CGA), a novel decentralized learning algorithm where (i) each agent aggregates cross-gradient information, i.e., derivatives of its model with respect to its neighbors' datasets, and (ii) updates its model using a projected gradient based on quadratic programming (QP). We theoretically analyze the convergence characteristics of CGA and demonstrate its efficiency on non-IID data distributions sampled from the MNIST and CIFAR-10 datasets. Our empirical comparisons show superior learning performance of CGA over existing state-of-the-art decentralized learning algorithms, as well as maintaining the improved performance under information compression to reduce peer-to-peer communication overhead. The code is available here on GitHub.