It is expected that the continued development and improvement of AI methods - by industry or university research - will make them even more powerful, and their demand in academic research will further grow. However, along with the progressive development of AI methods, their theoretical and algorithmic complexity as well as their requirements on practical implementations will grow to leverage their full potential. 

For these reasons, it is becoming increasingly difficult for domain scientists to keep pace with the latest developments in AI and apply the methods themselves. This divergence between the rapid development of AI methods and the complexity of their application on the one side and the high demand for AI applications in domain sciences on the other side has been recognized by universities and research institutes. Therefore, specialized units offering, e.g. machine learning support for domain scientists, have been established to fill this gap.

Nevertheless, even for expert data scientists, the application of AI methods to scientific research problems is fraught with challenges. Every scientific domain has established its unique language, standards and mindset. Additionally, when used in scientific domains, the behavior of the AI models must be constrained by domain-specific laws, e.g. physics-constrained ML applications, and choosing a suitable method is not always straightforward. The chosen methods must often be adapted or further developed to fit the specific use-cases.

With this track, we want to provide a platform for the stakeholders in such collaborations: data scientists, software engineers and domain scientists, and give them a chance to share their experiences. 

The speakers in this track will present specific technical and scientific projects they have worked on, highlighting one or more of the following topics:
a) Benefits and challenges of this collaboration model
b) Use cases of collaboration on specific projects
c) Challenges in applying machine learning to various scientific domains
d) Scaling up and professionalization of machine learning applications together with researchers
e) Opportunities for further improving this collaboration model

We strongly believe that this track will bring a lot of value and new input to those engaged in such collaborations and will further improve such synergies.