DADAO: Decoupled Accelerated Decentralized Asynchronous Optimization
This addresses efficient distributed optimization for networks, offering improvements in communication efficiency and computational speed.
The paper tackles decentralized optimization by introducing DADAO, a first-order algorithm that decouples computation and communication through independent Poisson processes, achieving accelerated rates for both local gradients and communications compared to state-of-the-art methods.
This work introduces DADAO: the first decentralized, accelerated, asynchronous, primal, first-order algorithm to minimize a sum of $L$-smooth and $μ$-strongly convex functions distributed over a given network of size $n$. Our key insight is based on modeling the local gradient updates and gossip communication procedures with separate independent Poisson Point Processes. This allows us to decouple the computation and communication steps, which can be run in parallel, while making the whole approach completely asynchronous. This leads to communication acceleration compared to synchronous approaches. Our new method employs primal gradients and does not use a multi-consensus inner loop nor other ad-hoc mechanisms such as Error Feedback, Gradient Tracking, or a Proximal operator. By relating the inverse of the smallest positive eigenvalue of the Laplacian matrix $χ_1$ and the maximal resistance $χ_2\leq χ_1$ of the graph to a sufficient minimal communication rate between the nodes of the network, we show that our algorithm requires $\mathcal{O}(n\sqrt{\frac{L}μ}\log(\frac{1}ε))$ local gradients and only $\mathcal{O}(n\sqrt{χ_1χ_2}\sqrt{\frac{L}μ}\log(\frac{1}ε))$ communications to reach a precision $ε$, up to logarithmic terms. Thus, we simultaneously obtain an accelerated rate for both computations and communications, leading to an improvement over state-of-the-art works, our simulations further validating the strength of our relatively unconstrained method.