We introduced a new framework to detect perceptual bugs using a Long
Short-Term Memory (LSTM) network, which detects bugs in video games as
anomalies. The detected buggy frames are then clustered to determine the
category of the occurred bug. The framework was evaluated on two First Person
Shooter (FPS) games. Results show the effectiveness of the framework.

我们提出了一个新的框架，使用长短期记忆（LSTM）网络来检测感知缺陷，将视频游戏中的缺陷检测为异常，并通过聚类来确定发生的缺陷类别。该框架在两个第一人称射击游戏上进行了评估，结果显示了框架的有效性。

使用 LSTM 网络自动检测游戏中的错误

Automatic Bug Detection in Games using LSTM Networks

Our work is at the crossroads of two categories of technologies. On the one
hand, omnichannel digit services, to address the needs of users in the most
seamless way. On the other hand, low code approaches, to build simply even
complex software applications. In this twofold context, we propose DSUL
(Digital Service Universal Language). It allows to build omnichannel services
with minimal work from their designers. We describe precisely how DSUL
operates, and its innovation in regard to the state of the art. We also
consider the various methods to evaluate this framework.

我们提出了一个名为 DSUL 的数字服务通用语言，它结合了全渠道数字服务和低代码方法，并能帮助设计者最小化工作量来构建数字服务。我们阐述了 DSUL 的操作方式及其相对于现有技术的创新，并探讨了不同的评估方法。