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 1812.09359cs.CL
收藏

NeuroX:分析神经网络中单个神经元的工具包

NeuroX: A Toolkit for Analyzing Individual Neurons in Neural Networks

Qatar Computing Research Institute, HBKU Research Complex(卡塔尔计算研究所,哈马德·本·哈利法大学研究中心);MIT Computer Science and Artificial Intelligence Laboratory(麻省理工学院计算机科学与人工智能实验室)

Fahim Dalvi, Avery Nortonsmith, D. Anthony Bau, Yonatan Belinkov, Hassan Sajjad, Nadir Durrani, James Glass

摘要:

我们提出了一个工具包,以促进解释和理解的神经网络模型。该工具包提供了几种方法来识别与模型本身或外部任务相关的显着神经元。用户可以可视化选定的神经元,消融它们以测量它们对模型准确性的影响,并操纵它们以控制测试时模型的行为。这样的分析有可能成为各种研究方向的跳板,例如理解模型,更好的架构选择,模型蒸馏和控制数据偏差。

英文摘要:

We present a toolkit to facilitate the interpretation and understanding of neural network models. The toolkit provides several methods to identify salient neurons with respect to the model itself or an external task. A user can visualize selected neurons, ablate them to measure their effect on the model accuracy, and manipulate them to control the behavior of the model at the test time. Such an analysis has a potential to serve as a springboard in various research directions, such as understanding the model, better architectural choices, model distillation and controlling data biases.

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。

↑