The reliability of software that has a Deep Neural Network (DNN) as a
component is urgently important today given the increasing number of critical
applications being deployed with DNNs. The need for reliability raises a need
for rigorous testing of the safety and trustworthiness of these systems. In the
last few years, there have been a number of research efforts focused on testing
DNNs. However the test generation techniques proposed so far lack a check to
determine whether the test inputs they are generating are valid, and thus
invalid inputs are produced. To illustrate this situation, we explored three
recent DNN testing techniques. Using deep generative model based input
validation, we show that all the three techniques generate significant number
of invalid test inputs. We further analyzed the test coverage achieved by the
test inputs generated by the DNN testing techniques and showed how invalid test
inputs can falsely inflate test coverage metrics.
To overcome the inclusion of invalid inputs in testing, we propose a
technique to incorporate the valid input space of the DNN model under test in
the test generation process. Our technique uses a deep generative model-based
algorithm to generate only valid inputs. Results of our empirical studies show
that our technique is effective in eliminating invalid tests and boosting the
number of valid test inputs generated.

针对系统安全和可信度需要进行深入测试的 DNN (DNN 指深度神经网络) 组件，提出了一种基于深度生成模型算法的有效测试方法，以消除无效的测试并提高测试输入的有效性。

使用生成模型进行神经网络的分布感知测试

Distribution-Aware Testing of Neural Networks Using Generative Models

This thesis explores a number of online machine learning algorithms. From a
theoret- ical perspective, it assesses their employability for a particular
function approximation problem where the analytical models fall short.
Furthermore, it discusses the applica- tion of theoretically suitable learning
algorithms to the function approximation problem at hand through an efficient
implementation that exploits various computational and mathematical shortcuts.
Finally, this thesis work evaluates the implemented learning algorithms
according to various evaluation criteria through rigorous testing.

本文探讨了多个在线机器学习算法，从理论角度评估了它们在特定函数逼近问题上的适用性，进一步通过高效的实现利用各种计算和数学技巧将理论上适合的学习算法应用于手头的函数逼近问题，并通过严格测试评估实现的学习算法。