Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2409.11995 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916399864086528 |
|---|---|
| author | Kiselev, Nikita Grabovoy, Andrey |
| author_facet | Kiselev, Nikita Grabovoy, Andrey |
| contents | The loss landscape of neural networks is a critical aspect of their training, and understanding its properties is essential for improving their performance. In this paper, we investigate how the loss surface changes when the sample size increases, a previously unexplored issue. We theoretically analyze the convergence of the loss landscape in a fully connected neural network and derive upper bounds for the difference in loss function values when adding a new object to the sample. Our empirical study confirms these results on various datasets, demonstrating the convergence of the loss function surface for image classification tasks. Our findings provide insights into the local geometry of neural loss landscapes and have implications for the development of sample size determination techniques. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_11995 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Unraveling the Hessian: A Key to Smooth Convergence in Loss Function Landscapes Kiselev, Nikita Grabovoy, Andrey Machine Learning The loss landscape of neural networks is a critical aspect of their training, and understanding its properties is essential for improving their performance. In this paper, we investigate how the loss surface changes when the sample size increases, a previously unexplored issue. We theoretically analyze the convergence of the loss landscape in a fully connected neural network and derive upper bounds for the difference in loss function values when adding a new object to the sample. Our empirical study confirms these results on various datasets, demonstrating the convergence of the loss function surface for image classification tasks. Our findings provide insights into the local geometry of neural loss landscapes and have implications for the development of sample size determination techniques. |
| title | Unraveling the Hessian: A Key to Smooth Convergence in Loss Function Landscapes |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2409.11995 |