Neural Network Plasticity and Loss Sharpness
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914957316063232 |
|---|---|
| author | Koster, Max Kukla, Jude |
| author_facet | Koster, Max Kukla, Jude |
| contents | In recent years, continual learning, a prediction setting in which the problem environment may evolve over time, has become an increasingly popular research field due to the framework's gearing towards complex, non-stationary objectives. Learning such objectives requires plasticity, or the ability of a neural network to adapt its predictions to a different task. Recent findings indicate that plasticity loss on new tasks is highly related to loss landscape sharpness in non-stationary RL frameworks. We explore the usage of sharpness regularization techniques, which seek out smooth minima and have been touted for their generalization capabilities in vanilla prediction settings, in efforts to combat plasticity loss. Our findings indicate that such techniques have no significant effect on reducing plasticity loss. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_17300 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Neural Network Plasticity and Loss Sharpness Koster, Max Kukla, Jude Machine Learning Artificial Intelligence In recent years, continual learning, a prediction setting in which the problem environment may evolve over time, has become an increasingly popular research field due to the framework's gearing towards complex, non-stationary objectives. Learning such objectives requires plasticity, or the ability of a neural network to adapt its predictions to a different task. Recent findings indicate that plasticity loss on new tasks is highly related to loss landscape sharpness in non-stationary RL frameworks. We explore the usage of sharpness regularization techniques, which seek out smooth minima and have been touted for their generalization capabilities in vanilla prediction settings, in efforts to combat plasticity loss. Our findings indicate that such techniques have no significant effect on reducing plasticity loss. |
| title | Neural Network Plasticity and Loss Sharpness |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2409.17300 |