تقرير
Higher-order mutual information reveals synergistic sub-networks for multi-neuron importance
| العنوان: | Higher-order mutual information reveals synergistic sub-networks for multi-neuron importance |
|---|---|
| المؤلفون: | Clauw, Kenzo, Stramaglia, Sebastiano, Marinazzo, Daniele |
| سنة النشر: | 2022 |
| المجموعة: | Computer Science Mathematics Quantitative Biology |
| مصطلحات موضوعية: | Computer Science - Machine Learning, Computer Science - Information Theory, Quantitative Biology - Neurons and Cognition |
| الوصف: | Quantifying which neurons are important with respect to the classification decision of a trained neural network is essential for understanding their inner workings. Previous work primarily attributed importance to individual neurons. In this work, we study which groups of neurons contain synergistic or redundant information using a multivariate mutual information method called the O-information. We observe the first layer is dominated by redundancy suggesting general shared features (i.e. detecting edges) while the last layer is dominated by synergy indicating local class-specific features (i.e. concepts). Finally, we show the O-information can be used for multi-neuron importance. This can be demonstrated by re-training a synergistic sub-network, which results in a minimal change in performance. These results suggest our method can be used for pruning and unsupervised representation learning. Comment: Paper presented at InfoCog @ NeurIPS 2022 |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2211.00416 |
| رقم الأكسشن: | edsarx.2211.00416 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |