Understandable science

On this page you will find a basic explanation of the main science topics I’m interested in. It is currently work in progress, so a lot of information is still missing. For a lengthy and somewhat more technical introduction to these topics you can also refer to chapters 1 to 4 of my master thesis.

The main interest of my research is the application of deep learning to the analysis of gravitational-wave data. I, therefore, focus on the application of neural networks to noisy time series data. Deep learning as a gravitational-wave data analysis tool is a topic that has started to gain momentum around the year 2017, when first publications showed that neural networks are capable of distinguishing simple gravitational-wave signals from simulated noise. Since then studies have started to explore more difficult signals, characterize noise, find noise anomalies, analyze the sources and many more tasks. Given that this line of research is new it needs to be thoroughly tested. Much of the work that I have conducted is to translate between machine learning algorithms and the established matched filter searches, where model waveforms are used to look for particular signals. We are trying to find applications of deep learning where their use is of actual benefit to the field and where it can perform tasks otherwise not possible. We also try to find limitations of the networks to explore new research areas.

When people looked at the speed of light, they noticed two things. Firstly, it is finite and secondly, it is the same irrespective of your own state of motion. Einstein used the idea that nothing can travel faster than light and that the speed of light is constant for any observer that is not accelerated as the basis for his theory of relativity. He also assumed that the description of the laws of physics in their most simple form does not depend on the speed of the observer describing these laws.

With these two basic principles the question of whether or not two events happen simultaneously is not easily answered. The only sensible definition of simultaneity is that two events happened at the same time when light emitted from these events arrive at the observer at the same time. Now consider an observer that lives on a straight line. Let’s call her Alice. She receives light from an event L1 on her left at the same time as she receives light from a second event R1 on her right. For her, these two events happened at the same time. Now consider Bob moving with a constant velocity from Alice’s left to her right side. He passes her, at the time where the light was sent out from L1 and R1 (as determined by Alice). However, Bob continues moving towards R1 and, therefore, the light from there will reach him before the light from L1 does. He is running away from L1 after all. But now the two light-rays from L1 and R1 don’t reach Bob at the same time anymore. Therefore, the two events are not happening at the same time for him.