Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.13130 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917409363853312 |
|---|---|
| author | Goyal, Saumya Rongali, Rohith Ray, Ritabrata Póczos, Barnabás |
| author_facet | Goyal, Saumya Rongali, Rohith Ray, Ritabrata Póczos, Barnabás |
| contents | We study learning to learn for regression problems through the lens of hyperparameter tuning. We propose the Langevin Gradient Descent Algorithm (LGD), which approximates the mean of the posterior distribution defined by the loss function and regularizer of a convex regression task. We prove the existence of an optimal hyperparameter configuration for which the LGD algorithm achieves the Bayes' optimal solution for squared loss. Subsequently, we study generalization guarantees on meta-learning optimal hyperparameters for the LGD algorithm from a given set of tasks in the data-driven setting. For a number of parameters $d$ and hyperparameter dimension $h$, we show a pseudo-dimension bound of $O(dh)$, upto logarithmic terms under mild assumptions on LGD. This matches the dimensional dependence of the bounds obtained in prior work for the elastic net, which only allows for $h=2$ hyperparameters, and extends their bounds to regression on convex loss. Finally, we show empirical evidence of the success of LGD and the meta-learning procedure for few-shot learning on linear regression using a few synthetically created datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_13130 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Generalization Guarantees on Data-Driven Tuning of Gradient Descent with Langevin Updates Goyal, Saumya Rongali, Rohith Ray, Ritabrata Póczos, Barnabás Machine Learning We study learning to learn for regression problems through the lens of hyperparameter tuning. We propose the Langevin Gradient Descent Algorithm (LGD), which approximates the mean of the posterior distribution defined by the loss function and regularizer of a convex regression task. We prove the existence of an optimal hyperparameter configuration for which the LGD algorithm achieves the Bayes' optimal solution for squared loss. Subsequently, we study generalization guarantees on meta-learning optimal hyperparameters for the LGD algorithm from a given set of tasks in the data-driven setting. For a number of parameters $d$ and hyperparameter dimension $h$, we show a pseudo-dimension bound of $O(dh)$, upto logarithmic terms under mild assumptions on LGD. This matches the dimensional dependence of the bounds obtained in prior work for the elastic net, which only allows for $h=2$ hyperparameters, and extends their bounds to regression on convex loss. Finally, we show empirical evidence of the success of LGD and the meta-learning procedure for few-shot learning on linear regression using a few synthetically created datasets. |
| title | Generalization Guarantees on Data-Driven Tuning of Gradient Descent with Langevin Updates |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2604.13130 |