Quantized Decentralized Consensus Optimization
Quantized Decentralized Consensus Optimization
复制标题
DOI：
10.1109/cdc.2018.8619539
复制
发表时间：
2018-06
期刊：
影响因子：
--
通讯作者：
Amirhossein Reisizadeh;Aryan Mokhtari;S. Hassani;Ramtin Pedarsani
中科院分区：
文献类型：
--
作者：
Amirhossein Reisizadeh;Aryan Mokhtari;S. Hassani;Ramtin Pedarsani
我们考虑分散共识优化的问题，其中$ n $ coSevex函数的总和比构成连接网络的$ n $分布式代理的总和最小化。特别是，我们考虑了以下情况：为了减轻分布式优化中的通信瓶颈，节点之间传达的本地决策变量进行了量化。我们提出了量化的分散梯度下降（QDGD）算法，其中节点通过将从邻居接收到的量化信息与本地信息相结合来更新其本地决策变量。我们证明，在标准的强凸度和局部成本功能的平稳性假设下，QDGD实现了消失的平均解决方案误差。据我们所知，这是第一种在存在量化噪声的情况下达到共识误差的第一种算法。此外，我们提供的模拟结果表明我们派生的理论收敛率与实验结果之间的一致性很紧。
We consider the problem of decentralized consensus optimization, where the sum of $n$ convex functions are minimized over $n$ distributed agents that form a connected network. In particular, we consider the case that the communicated local decision variables among nodes are quantized in order to alleviate the communication bottleneck in distributed optimization. We propose the Quantized Decentralized Gradient Descent (QDGD) algorithm, in which nodes update their local decision variables by combining the quantized information received from their neighbors with their local information. We prove that under standard strong convexity and smoothness assumptions for local cost functions, QDGD achieves a vanishing mean solution error. To the best of our knowledge, this is the first algorithm that achieves vanishing consensus error in the presence of quantization noise. Moreover, we provide simulation results that show tight agreement between our derived theoretical convergence rate and the experimental results.