Please note:
On this page you will only see the English-language presentations of the conference. You can find all conference sessions, including the German speaking ones, here.
The times given in the conference program of OOP 2023 Digital correspond to Central European Time (CET).
By clicking on "VORTRAG MERKEN" within the lecture descriptions you can arrange your own schedule. You can view your schedule at any time using the icon in the upper right corner.
Testing AI Systems
At first glance, testing AI systems seems very different from testing “conventional” systems. However, many standard testing activities can be preserved as they are or only need small extensions.
In this talk, we give an overview of topics that will help you test AI systems: Attributes of training/testing/validation data, model performance metrics, and the statistical nature of AI systems. We will then relate these to testing tasks and show you how to integrate them.
Target Audience: Developers, Testers, Architects
Prerequisites: Basic knowledge of testing
Level: Basic
Extended Abstract:
From a testing perspective, systems that include AI components seem like a nightmare at first glance. How can you test a system that contains enough math to fill several textbooks and changes its behavior on the whims of its input data? How can you test what even its creators don’t fully understand?
Keep calm, grab a towel and carry on - what you have already been doing is still applicable, and most of the new things you should know are not as arcane as they might seem. Granted, some dimensions of AI systems like bias or explainability will likely not be able to be tested for in all cases. However, this complexity has been around for decades even in systems without any AI whatsoever. Additionally, you will have allies: Data scientists love talking about testing data.
In this talk, we give an overview of topics that will help you test AI systems: Attributes of training/testing/validation data, model performance metrics, and the statistical nature of AI systems. We will then relate these to testing tasks and show you how to integrate them.
Gregor Endler holds a doctor's degree in Computer Science for his thesis on completeness estimation of timestamped data. His work at Codemanufaktur GmbH deals with Machine Learning and Data Analysis.
Marco Achtziger is a Test Architect working for Siemens Healthcare GmbH in Forchheim. He has several qualifications from iSTQB and iSQI and is a certified Software Architect by Siemens AG.
Vortrag Teilen