stochastic gradient descent (SGD) is one of the most widely used optimization
methods for parallel and distributed processing of large datasets. One of the
key limitations of distributed SGD is the need to regula
我们扩展了 Approximate-Proximal Point 方法,在随机凸优化问题中应用包括随机次梯度、近端点和束方法,同时提出了更快的模型算法和加速方案,保持了 Approximate-Proximal Point 算法的鲁棒性,同时提供了更快的收敛速度和更低的界限。我们通过实证测试证实了理论结果的可行性。