Contributed Talk - Splinter Learning
Tuesday, 22 September 2020, 11:15 (virtual room G)
MEGAVIS - Real-time spectra analysis and visualization with autoencoders
Antonio D'Isanto, Kai Polsterer
The data explosion in astronomy requires the development of new techniques both from the infrastructure and from the analysis side. In particular, the increase of the data complexity demands a parallel effort to deliver efficient and standardized solutions for accessing and managing data, tools and software. To this purpose is devoted the ESCAPE project, which aims to build a huge European collaboration to face the new challenges given by data-driven research, complex data workflows, infrastractural issues, data and software interoperability. The final goal is to deliver a new platform which could extend the concept of VO, combining data, software and expertise in an open-source repository that looks toward the approaching exabyte-era. In this talk I will give an overview of the work fulfilled within the project, presenting MEGAVIS, a prototype based on machine learning, which aims to start building a new paradigm for data access and search, not based on explicit criteria but implicitly, looking at similarities. The prototype is based on dimensionality reduction models, and in particular on an autoencoder. A demo will be shown to illustrate the main features and capabilities of the software, and the possibilities of future developments.