论文标题
黑匣子到白色框：基于战略探测的模型特征
Black Box to White Box: Discover Model Characteristics Based on Strategic Probing
论文作者
论文摘要
在机器学习中，白盒对抗攻击依赖于了解有关模型属性的基本知识。这项工作的重点是发现分散模型信息的部分：基础架构和主要培训数据集。随着本文的过程，一组结构化的输入探针和模型的输出成为深层分类器的训练数据。探索了机器学习中的两个子域：带有GPT-2的基于图像的分类器和文本变压器。通过图像分类，重点是探索流行公共库中可用的常见部署架构和数据集。使用具有多个参数级别的单个变压器体系结构，通过微调不同的数据集来探索文本生成。图像和文本中探索的每个数据集都可以彼此区分。文本变压器输出中的多样性意味着需要进一步的研究来成功地对文本域中的体系结构归因进行分类。
In Machine Learning, White Box Adversarial Attacks rely on knowing underlying knowledge about the model attributes. This works focuses on discovering to distrinct pieces of model information: the underlying architecture and primary training dataset. With the process in this paper, a structured set of input probes and the output of the model become the training data for a deep classifier. Two subdomains in Machine Learning are explored: image based classifiers and text transformers with GPT-2. With image classification, the focus is on exploring commonly deployed architectures and datasets available in popular public libraries. Using a single transformer architecture with multiple levels of parameters, text generation is explored by fine tuning off different datasets. Each dataset explored in image and text are distinguishable from one another. Diversity in text transformer outputs implies further research is needed to successfully classify architecture attribution in text domain.
