Convergence Analysis of Optimization Algorithms
HyoungSeok Kim, JiHoon Kang, WooMyoung Park, SukHyun Ko, YoonHo Cho, DaeSung Yu, YoungSook Song, JungWon Choi
arXiv:1707.01647v18 citations
Originality Synthesis-oriented
AI Analysis
This provides incremental guidance for researchers and practitioners in machine learning on optimizer choice.
The paper analyzes regret bounds of optimization algorithms to guide algorithm selection based on dataset and loss function, assuming convex and Lipschitz continuous loss functions.
The regret bound of an optimization algorithms is one of the basic criteria for evaluating the performance of the given algorithm. By inspecting the differences between the regret bounds of traditional algorithms and adaptive one, we provide a guide for choosing an optimizer with respect to the given data set and the loss function. For analysis, we assume that the loss function is convex and its gradient is Lipschitz continuous.