30 December 2020 in IEEE Transactions on Signal Processing
The emergence of big data has caused a dramatic shift in the operating regime for optimization algorithms. The performance bottleneck, which used to be computations, is now often communications. Several gradient compression techniques have been proposed to reduce the communication load at the price of a loss in solution accuracy. Recently, it has been shown how compression errors can be compensated for in the optimization algorithm to improve the solution accuracy. Even though convergence guarantees for error-compensated algorithms have been established, there is very limited theoretical support for quantifying the observed improvements in solution accuracy. In this paper, we show that Hessian-aided error compensation, unlike other existing schemes, avoids accumulation of compression errors on quadratic problems. We also present strong convergence guarantees of Hessian-based error compensation for stochastic gradient descent. Our numerical experiments highlight the benefits of Hessian-based error compensation, and demonstrate that similar convergence improvements are attained when only a diagonal Hessian approximation is used.
26 January 2023 in IEEE Transactions on Signal Processing
Noisy gradient algorithms have emerged as one of the most popular algorithms for distributed optimization with massive data. Choosing proper step-size schedules is an important task to tune in the algorithms for good performance. For the algorithms to attain fast convergence and high accuracy, it is intuitive to use large step-sizes in the initial iterations when the gradient noise is typically small compared to the algorithm-steps, and reduce the step-sizes as the algorithm progresses. This intuition has been confirmed in theory and practice for stochastic gradient descent. However, similar results are lacking for other methods using approximate gradients. This paper shows that the diminishing step-size strategies can indeed be applied for a broad class of noisy gradient algorithms. Our analysis framework is based on two classes of systems that characterize the impact of the step-sizes on the convergence performance of many algorithms. Our results show that such step-size schedules enable these algorithms to enjoy the optimal rate. We exemplify our results on stochastic compression algorithms. Our experiments validate fast convergence of these algorithms with the step decay schedules.
5 July 2024 in IEEE Transactions on Big Data
Data similarity assumptions have traditionally been relied upon to understand the convergence behaviors of federated learning methods. Unfortunately, this approach often demands fine-tuning step sizes based on the level of data similarity. When data similarity is low, these small step sizes result in an unacceptably slow convergence speed for federated methods. In this paper, we present a novel and unified framework for analyzing the convergence of federated learning algorithms without the need for data similarity conditions. Our analysis centers on an inequality that captures the influence of step sizes on algorithmic convergence performance. By applying our theorems to well-known federated algorithms, we derive precise expressions for three widely used step size schedules: fixed, diminishing, and step-decay step sizes, which are independent of data similarity conditions. Finally, we conduct comprehensive evaluations of the performance of these federated learning algorithms, employing the proposed step size strategies to train deep neural network models on benchmark datasets under varying data similarity conditions. Our findings demonstrate significant improvements in convergence speed and overall performance, marking a substantial advancement in federated learning research.
8 January 2025 in IEEE Transactions on Control of Network Systems
Parallel stochastic gradient methods are gaining prominence in solving large-scale machine learning problems that involve data distributed across multiple nodes. However, obtaining unbiased stochastic gradients, which have been the focus of most theoretical research, is challenging in many distributed machine learning applications. The gradient estimations easily become biased, for example, when gradients are compressed or clipped, when data is shuffled, and in meta-learning and reinforcement learning. In this work, we establish worst-case bounds on parallel momentum methods under biased gradient estimation on both general non-convex and $\mu$ -PL problems. Our analysis covers general distributed optimization problems, and we work out the implications for special cases where gradient estimates are biased, i.e. in meta-learning and when the gradients are compressed or clipped. Our numerical experiments verify our theoretical findings and show faster convergence performance of momentum methods than traditional biased gradient descent.