New software based on Artificial Intelligence helps to interpret complex data
Experimental data is often not only highly dimensional, but also noisy and full of artefacts. This makes it difficult to interpret the data. Now a team at HZB has designed software that uses self-learning neural networks to compress the data in a smart way and reconstruct a low-noise version in the next step. This enables to recognise correlations that would otherwise not be discernible. The software has now been successfully used in photon diagnostics at the FLASH free electron laser at DESY. But it is suitable for very different applications in science.
More is not always better, but sometimes a problem. With highly complex data, which have many dimensions due to their numerous parameters, correlations are often no longer recognisable. Especially since experimentally obtained data are additionally disturbed and noisy due to influences that cannot be controlled.
Helping humans to interpret the data
Now, new software based on artificial intelligence methods can help: It is a special class of neural networks (NN) that experts call "disentangled variational autoencoder network (β-VAE)". Put simply, the first NN takes care of compressing the data, while the second NN subsequently reconstructs the data. "In the process, the two NNs are trained so that the compressed form can be interpreted by humans," explains Dr Gregor Hartmann. The physicist and data scientist supervises the Joint Lab on Artificial Intelligence Methods at HZB, which is run by HZB together with the University of Kassel.
Extracting core principles without prior knowledge
Google Deepmind had already proposed to use β-VAEs in 2017. Many experts assumed that the application in the real world would be challenging, as non-linear components are difficult to disentangle. "After several years of learning how the NNs learn, it finally worked," says Hartmann. β-VAEs are able to extract the underlying core principle from data without prior knowledge.
Photon energy of FLASH determined
In the study now published, the group used the software to determine the photon energy of FLASH from single-shot photoelectron spectra. "We succeeded in extracting this information from noisy electron time-of-flight data, and much better than with conventional analysis methods," says Hartmann. Even data with detector-specific artefacts can be cleaned up this way.
A powerful tool for different problems
"The method is really good when it comes to impaired data," Hartmann emphasises. The programme is even able to reconstruct tiny signals that were not visible in the raw data. Such networks can help uncover unexpected physical effects or correlations in large experimental data sets. "AI-based intelligent data compression is a very powerful tool, not only in photon science," says Hartmann.
Now plug and play
In total, Hartmann and his team spent three years developing the software. "But now, it is more or less plug and play. We hope that soon many colleagues will come with their data and we can support them."
- Peat as a sustainable precursor for fuel cell catalyst materialsIron-nitrogen-carbon catalysts have the potential to replace the more expensive platinum catalysts currently used in fuel cells. This is shown by a study conducted by researchers from the Helmholtz-Zentrum Berlin (HZB), Physikalisch-Technische Bundesanstalt (PTB) and universities in Tartu and Tallinn, Estonia. At BESSY II, the team observed the formation of complex microstructures within various samples. They then analysed which structural parameters were particularly important for fostering the preferred electrochemical reactions. The raw material for such catalysts is well decomposed peat.
- Helmholtz Investigator Group on magnonsDr Hebatalla Elnaggar is setting up a new Helmholtz Investigator Group at HZB. At BESSY II, the materials scientist will investigate so-called magnons in magnetic perovskite thin films. The aim is to lay the foundations for future terahertz magnon technology: magnonic devices operating in the terahertz range could process data using a fraction of the energy required by the most advanced semiconductor devices, and at speeds up to a thousand times faster.
- The future of corals – what X-rays can tell usThis summer, it was all over the media. Driven by the climate crisis, the oceans have now also passed a critical point, the absorption of CO2 is making the oceans increasingly acidic. The shells of certain sea snails are already showing the first signs of damage. But also the skeleton structures of coral reefs are deteriorating in more acidic conditions. This is especially concerning given that corals are already suffering from marine heatwaves and pollution, which are leading to bleaching and finally to the death of entire reefs worldwide. But how exactly does ocean acidification affect reef structures?
Prof. Dr. Tali Mass, a marine biologist from the University of Haifa, Israel, is an expert on stony corals. Together with Prof. Dr. Paul Zaslansky, X-ray imaging expert from Charité Berlin, she investigated at BESSY II the skeleton formation in baby corals, raised under different pH conditions. Antonia Rötger spoke online with the two experts about the results of their recent study and the future of coral reefs.