The workshop takes place at the Princeton University Campus, Mader Hall.
The workshop aims at bringing together the fusion community to discuss the challenges brought by AI and visualizing large datasets in fusion experiment and simulation.
The three day in hybrid mode event focuses on feedback, lessons learnt and innovative techniques that developers and users experience with AI techniques and visualizing large datasets. A summary of the workshop with all contributors for a journal is one of the key outcomes.
The workshop takes place at the Princeton University Campus, Mader Hall.
Deadline for abstract submission : 24th May 2024.
Main topics:
- Topic 1 :Smart indexing of dormant data: How to visualize large signals
- Retrieving all the points of the signal is impossible. Decimation, downsampling techniques are cruciaL
- Place of offline processing : client side versus server side, how to properly downsample a signal [ traditional min/max/avg may not be enough]
- How to handle dynamic processing of large signals?
- Is real-time processing of the raw signal the direction?
- Place of AI in downsampling techniques and integrated data analysis
- Topic 2: the use of AI as part of operations
- how to trust the AI workflow?
- how to validate and verify the workflow?
- place of security?
- Topic 3 the use of cloud systems in fusion. More and more facilities are using cloud in their environment, either to store and/or compute data
- How cloud is impacting data visualization? Shall everything run in the cloud? Web ?
- Computing power ?
- Open data?
- Topic 3: the place of open source utilities: there are very few open sources data visualization tools in fusion?
- Maybe we need a common github entry point where one can find all the open sources related to
- Language? Python, R?
Registration
Registration for this event is currently open.