Deploying a Self-Supervised Learning Based Model to Search Events Across Space and Time

Motivation
- Scientific Study of natural events, phenomena, or disasters require examples which span across time and space.
- Machine Learning adaptation is on the rise, but there’s a lack of labeled training datasets that could be used to train
or validate the models.
- Best case scenario:
- There’s an event database that tracks events available through time and space.
- Provides all data associated with the events.
- Real life scenario:
- Some events are better tracked than others.
- Scientists need to spend significant time identifying and gathering examples of events from different sources.