Database management systems (DBMS) in the cloud are the backbone for managing large volumes of data efficiently and thus play a central role in business and science today. For providing high performance, many of the most complex DBMS components such as query optimizers or schedulers involve solving non-trivial problems.
To tackle such problems, very recent work has outlined a new direction of so-called learned DBMS components where AI-based methods are used to replace and enhance core DBMS components which has shown to provide significant performance benefits. This route is in particular interesting since Cloud vendors such as Google, Amazon, and Microsoft are already applying these techniques to optimize the performance their cloud data systems.
Furthermore, AI has also been used for improving many other data management related tasks such as data engineering tasks (e.g., error detection and correction in databases or data transformation and data augmentation) which typically cause high manual overhead and can be automated by the use of AI. Finally, AI has also been used for extending databases by better data access interfaces (e.g., natural languague querying and chatbots for data) or by supporting data beyond structured tabular data (i.e., text and images).
This seminar serves the purpose to understand the basic concepts of how AI can be used for data management. In the first part of the seminar, participants will learn the basics of AI for data management along with implementing a case study themselves. In the second part, every participant will select and present a recent research paper. The papers will typically be recent publications in relevant research venues and journals such as SIGMOD, VLDB or ICML, NeurIPS.
The seminar kick-off will typically be in the first two weeks of the semester where we discuss the organization of the seminar. Further information can be found at: http://tuda.systems
- Dozent*in: Jan-Micha Bodensohn