arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1612.09438cs.LG
收藏

基于神经元群关联先验的物理和语义概念自动发现

Automatic Discoveries of Physical and Semantic Concepts via Association Priors of Neuron Groups

Chinese University of Hong Kong(香港中文大学);South China University of Technology(华南理工大学)

Shuai Li, Kui Jia, Xiaogang Wang

摘要:

最近成功的深度神经网络在很大程度上是以监督的方式训练的。它{ \it}将输入样本的复杂模式与最后一层中的神经元相关联,这些神经元形成{ \it概念}的表示。尽管他们取得了成功,但与学习概念相关的复杂模式的特性仍然难以捉摸。在这项工作中,通过分析神经元如何与监督网络中的概念相关联,我们假设有适当的先验来调节学习,神经网络可以自动将中间层的神经元与与现实世界概念相关联的概念相关联,当只使用将概念与顶级神经元相关联的标签进行训练时,这是无监督学习的一种合理方式。我们开发了一个先验来验证假设,并通过实验发现所提出的先验帮助神经网络自动学习较低层的两个基本物理概念,例如,过滤器的旋转,以及更高层的高度语义概念,例如,入门级类别的细粒度类别。

英文摘要:

The recent successful deep neural networks are largely trained in a supervised manner. It {\it associates} complex patterns of input samples with neurons in the last layer, which form representations of {\it concepts}. In spite of their successes, the properties of complex patterns associated a learned concept remain elusive. In this work, by analyzing how neurons are associated with concepts in supervised networks, we hypothesize that with proper priors to regulate learning, neural networks can automatically associate neurons in the intermediate layers with concepts that are aligned with real world concepts, when trained only with labels that associate concepts with top level neurons, which is a plausible way for unsupervised learning. We develop a prior to verify the hypothesis and experimentally find the proposed prior help neural networks automatically learn both basic physical concepts at the lower layers, e.g., rotation of filters, and highly semantic concepts at the higher layers, e.g., fine-grained categories of an entry-level category.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑