Understanding the algorithmic regularization effect of stochastic gradient descent (SGD) is one of the key challenges in modern machine learning and deep learning theory. Most of the existing works, however, focus on very small or even infinitesimal learning rate regime, and fail to cover practical scenarios where the learning rate is moderate and annealing. In this paper, we make an initial attempt to characterize the particular regularization effect of SGD in the moderate learning rate regime by studying its behavior for optimizing an overparameterized linear regression problem. In this case, SGD and GD are known to converge to the unique minimum-norm solution; however, with the moderate and annealing learning rate, we show that they exhibit different directional bias: SGD converges along the large eigenvalue directions of the data matrix, while GD goes after the small eigenvalue directions. Furthermore, we show that such directional bias does matter when early stopping is adopted, where the SGD output is nearly optimal but the GD output is suboptimal. Finally, our theory explains several folk arts in practice used for SGD hyperparameter tuning, such as (1) linearly scaling the initial learning rate with batch size; and (2) overrunning SGD with high learning rate even when the loss stops decreasing.

Oja's algorithm has been the cornerstone of streaming methods in Principal Component Analysis (PCA) since it was first proposed in 1982. However, Oja's algorithm does not have a standardized choice of learning rate (step size) that both performs well in practice and truly conforms to the online streaming setting. In this paper, we propose a new learning rate scheme for Oja's method called AdaOja. This new algorithm requires only a single pass over the data and does not depend on knowing properties of the data set a priori. AdaOja is a novel variation of the Adagrad algorithm to Oja's algorithm in the single eigenvector case and extended to the multiple eigenvector case. We demonstrate for dense synthetic data, sparse real-world data and dense real-world data that AdaOja outperforms common learning rate choices for Oja's method. We also show that AdaOja performs comparably to state-of-the-art algorithms (History PCA and Streaming Power Method) in the same streaming PCA setting.

We propose a novel diminishing learning rate scheme, coined Decreasing-Trend-Nature (DTN), which allows us to prove fast convergence of the Stochastic Gradient Descent (SGD) algorithm to a first-order stationary point for smooth general convex and some class of nonconvex including neural network applications for classification problems. We are the first to prove that SGD with diminishing learning rate achieves a convergence rate of $\mathcal{O}(1/t)$ for these problems. Our theory applies to neural network applications for classification problems in a straightforward way.