用 Dirichlet Process Gaussian Mixture Model 分析 Convnet 学到的特征
Analyzing Learned Convnet Features with Dirichlet Process Gaussian Mixture Models
AI 导读
一篇 arXiv 论文提出用 Dirichlet Process Gaussian Mixture Model 对卷积神经网络(Convnet)各层所有节点的内部表征进行聚类,以可视化其学到的特征表示,应对内部表征高维度和深层高抽象带来的可视化难题。
正文 · 原文
Abstract:Convolutional Neural Networks (Convnets) have achieved good results in a range of computer vision tasks the recent years. Though given a lot of attention, visualizing the learned representations to interpret Convnets, still remains a challenging task. The high dimensionality of internal representations and the high abstractions of deep layers are the main challenges when visualizing Convnet functionality. We present in this paper a technique based on clustering internal Convnet representations with a Dirichlet Process Gaussian Mixture Model, for visualization of learned representations in Convnets. Our method copes with the high dimensionality of a Convnet by clustering representations across all nodes of each layer. We will discuss how this application is useful when considering transfer learning, i.e.\ transferring a model trained on one dataset to solve a task on a different one.
| Comments: | Presented at NIPS 2016 Workshop: Practical Bayesian Nonparametrics |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:1702.07189 [cs.CV] |
| (or arXiv:1702.07189v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.1702.07189 arXiv-issued DOI via DataCite |
Submission history
From: David Malmgren-Hansen Mr. [view email]
[v1]
Thu, 23 Feb 2017 12:11:42 UTC (3,633 KB)
来源:arXiv cs.AI · arxiv.org