High Probability Convergence Bounds for Non-convex Stochastic Gradient Descent with Sub-Weibull Noise
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2020
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911278929281024 |
|---|---|
| author | Madden, Liam Dall'Anese, Emiliano Becker, Stephen |
| author_facet | Madden, Liam Dall'Anese, Emiliano Becker, Stephen |
| contents | Stochastic gradient descent is one of the most common iterative algorithms used in machine learning and its convergence analysis is a rich area of research. Understanding its convergence properties can help inform what modifications of it to use in different settings. However, most theoretical results either assume convexity or only provide convergence results in mean. This paper, on the other hand, proves convergence bounds in high probability without assuming convexity. Assuming strong smoothness, we prove high probability convergence bounds in two settings: (1) assuming the Polyak-Łojasiewicz inequality and norm sub-Gaussian gradient noise and (2) assuming norm sub-Weibull gradient noise. In the second setting, as an intermediate step to proving convergence, we prove a sub-Weibull martingale difference sequence self-normalized concentration inequality of independent interest. It extends Freedman-type concentration beyond the sub-exponential threshold to heavier-tailed martingale difference sequences. We also provide a post-processing method that picks a single iterate with a provable convergence guarantee as opposed to the usual bound for the unknown best iterate. Our convergence result for sub-Weibull noise extends the regime where stochastic gradient descent has equal or better convergence guarantees than stochastic gradient descent with modifications such as clipping, momentum, and normalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2006_05610 |
| institution | arXiv |
| publishDate | 2020 |
| record_format | arxiv |
| spellingShingle | High Probability Convergence Bounds for Non-convex Stochastic Gradient Descent with Sub-Weibull Noise Madden, Liam Dall'Anese, Emiliano Becker, Stephen Optimization and Control Stochastic gradient descent is one of the most common iterative algorithms used in machine learning and its convergence analysis is a rich area of research. Understanding its convergence properties can help inform what modifications of it to use in different settings. However, most theoretical results either assume convexity or only provide convergence results in mean. This paper, on the other hand, proves convergence bounds in high probability without assuming convexity. Assuming strong smoothness, we prove high probability convergence bounds in two settings: (1) assuming the Polyak-Łojasiewicz inequality and norm sub-Gaussian gradient noise and (2) assuming norm sub-Weibull gradient noise. In the second setting, as an intermediate step to proving convergence, we prove a sub-Weibull martingale difference sequence self-normalized concentration inequality of independent interest. It extends Freedman-type concentration beyond the sub-exponential threshold to heavier-tailed martingale difference sequences. We also provide a post-processing method that picks a single iterate with a provable convergence guarantee as opposed to the usual bound for the unknown best iterate. Our convergence result for sub-Weibull noise extends the regime where stochastic gradient descent has equal or better convergence guarantees than stochastic gradient descent with modifications such as clipping, momentum, and normalization. |
| title | High Probability Convergence Bounds for Non-convex Stochastic Gradient Descent with Sub-Weibull Noise |
| topic | Optimization and Control |
| url | https://arxiv.org/abs/2006.05610 |