We consider nonconvex stochastic optimization problems in the asynchronous centralized distributed setup where the communication times from workers to a server can not be ignored, and the computation and communication times are potentially different for all workers.

We study the asynchronous stochastic gradient descent algorithm, for distributed training over $n$ workers that might be heterogeneous. In this algorithm, workers compute stochastic gradients in parallel at their own pace and return them to the server without any synchronization.Existing convergence rates of this algorithm for non-convex smooth objectives depend on the maximum delay $\tau_{\max}$ and reach an $\epsilon$-stationary point after $O\!\left(\sigma^2\epsilon^{-2}+ \tau_{\max}\epsilon^{-1}\right)$ iterations, where $\sigma$ is the variance of stochastic gradients.

Plain parallel SGD still faces many challenges in practice, motivat-CISP A Helmholtz Center for Information Security 36th Conference on Neural Information Processing Systems (NeurIPS 2022).

We consider nonconvex stochastic optimization problems in the asynchronous centralized distributed setup where the communication times from workers to a server can not be ignored, and the computation and communication times are potentially different for all workers.

"virtual iterates" and delay-adaptive stepsizes, which allow us to derive state-of-the-art guarantees for both convex and non-convex objectives.