Timezone: »

 
Poster
Asynchronous Decentralized Online Learning
Jiyan Jiang · Wenpeng Zhang · Jinjie GU · Wenwu Zhu

Thu Dec 09 12:30 AM -- 02:00 AM (PST) @
in Poster Session 5 »

Most existing algorithms in decentralized online learning are conducted in the synchronous setting. However, synchronization makes these algorithms suffer from the straggler problem, i.e., fast learners have to wait for slow learners, which significantly reduces such algorithms' overall efficiency. To overcome this problem, we study decentralized online learning in the asynchronous setting, which allows different learners to work at their own pace. We first formulate the framework of Asynchronous Decentralized Online Convex Optimization, which specifies the whole process of asynchronous decentralized online learning using a sophisticated event indexing system. Then we propose the Asynchronous Decentralized Online Gradient-Push (AD-OGP) algorithm, which performs asymmetric gossiping communication and instantaneous model averaging. We further derive a regret bound of AD-OGP, which is a function of the network topology, the levels of processing delays, and the levels of communication delays. Extensive experiments show that AD-OGP runs significantly faster than its synchronous counterpart and also verify the theoretical results.

Keywords: Optimization · Online Learning

Author Information

Jiyan Jiang (Tsinghua University)
Wenpeng Zhang (Ant Group)
Jinjie GU (Ant Group)
Wenwu Zhu (Tsinghua University)

More from the Same Authors