تقرير
Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets
| العنوان: | Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets |
|---|---|
| المؤلفون: | Power, Alethea, Burda, Yuri, Edwards, Harri, Babuschkin, Igor, Misra, Vedant |
| سنة النشر: | 2022 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Machine Learning |
| الوصف: | In this paper we propose to study generalization of neural networks on small algorithmically generated datasets. In this setting, questions about data efficiency, memorization, generalization, and speed of learning can be studied in great detail. In some situations we show that neural networks learn through a process of "grokking" a pattern in the data, improving generalization performance from random chance level to perfect generalization, and that this improvement in generalization can happen well past the point of overfitting. We also study generalization as a function of dataset size and find that smaller datasets require increasing amounts of optimization for generalization. We argue that these datasets provide a fertile ground for studying a poorly understood aspect of deep learning: generalization of overparametrized neural networks beyond memorization of the finite training dataset. Comment: Correspondence to alethea@openai.com. Code available at: https://github.com/openai/grok |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2201.02177 |
| رقم الأكسشن: | edsarx.2201.02177 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |