arXiv:2606. 01128v1 Announce Type: new Abstract: Communication overhead is a crucial bottleneck in scalable distributed learning.
arXiv:2405. 11667v2 Announce Type: replace Abstract: Local SGD is a popular optimization method in distributed learning, often outperforming other algorithms in practice, including mini-batch SGD.
arXiv:2608. 06563v1 Announce Type: new Abstract: Machine learning and optimization have advanced together, with practical demands motivating new theory and theoretical breakthroughs enabling new applications.
arXiv:2302. 09832v4 Announce Type: replace Abstract: In distributed optimization and federated learning, slow and costly communication between parallel devices and the central server constitutes the primary bottleneck.
arXiv:2606. 11081v1 Announce Type: cross Abstract: Communication-efficient pre-training of LLMs is increasingly important as training draws on compute distributed across clusters, data centers, and lower-bandwidth links.
arXiv:2607. 03011v1 Announce Type: cross Abstract: Model merging techniques, which aggregate independently finetuned models into one to combine their capabilities, have become a topic of significant interest in recent years, with a broad array of methods having been proposed to tackle this problem.
Communication-efficient pre-training of LLMs is increasingly important as training draws on compute distributed across clusters, data centers, and lower-bandwidth links. Many practical methods reduce communication frequency but still rely on synchronous All-Reduce operations that maintain identical model states and tie progress to global collectives.
arXiv:2604. 24012v3 Announce Type: replace Abstract: Federated learning enables a population of clients to collaboratively train machine learning models without exchanging their raw data, but standard algorithms such as FedAvg suffer from slow convergence and high communication and memory costs in heterogeneous, resource-constrained environments.
arXiv:2606. 07496v1 Announce Type: new Abstract: Decentralized stochastic optimization is a fundamental paradigm for large-scale learning over networks, where agents communicate only with their neighbors and no central coordinator is required.
The paper introduces Batched SGD, a variant that groups online samples into epochs and performs a single update per epoch using a low‑variance gradient estimate. This batching approach allows a straightforward high‑probability analysis without restrictive assumptions or auxiliary sequences, yielding near‑optimal rates for both strongly convex and non‑convex objectives under standard smoothness and sub‑Gaussian noise conditions. The authors also extend the method to federated learning, providing the first high‑probability guarantees with logarithmic communication complexity, linear speedup in the number of agents, and robustness to data heterogeneity.
FedLore introduces a communication- and memory-efficient federated learning framework that shares a low-rank optimization basis across clients each round, mitigating subspace fragmentation and enabling exact low-rank aggregation. By refreshing this shared basis across rounds, FedLore allows model updates to exceed the per-round rank budget while maintaining a provable $O(T^{-1/2})$ stationarity bound under standard assumptions. Experiments on vision and language tasks, including federated pre‑training, demonstrate that FedLore outperforms low‑rank adapter baselines and matches or surpasses full‑parameter training while reducing communication and optimizer‑state memory.
The paper addresses bias introduced by aggregating local signs in distributed sign-based variance reduction methods, which hampers optimal convergence rates. By proposing an unbiased compression of recursive gradient increments to track the global gradient at the server, the authors achieve optimal convergence rates for both nonconvex stochastic and finite-sum optimization. They provide specific rate bounds for α-norms and demonstrate matching sample complexities to centralized settings for finite-sum problems.
arXiv:2607. 01678v1 Announce Type: new Abstract: Communication increasingly dominates the cost of Large Language Model (LLM) pre-training, especially under data-parallel and sharded training schemes, where gradient synchronization and parameter reconstruction overhead increase with model size and system scale.