This episode is brought to you by the Cloud Wars Expo. This in-person event will be held from June 28th to 30th at the Moscone Center in San Francisco, California.
Highlights
00:35 — In an effort to create early warning systems, predicting earthquakes has been a constant issue for scientists.
01:26 — There has been a recent breakthrough for researchers in France who have created a machine learning model to pinpoint the location and size of a strong earthquake. It detects adjustments to Earth’s gravitational field that move during a massive earthquake.
02:02 — The researchers created a model called PEGSNet, which stands for Prompt ElastoGravity Signals. The model has been trained with a combination of seismic data collected in Japan along with 500,000 simulated gravity signals for earthquakes.
02:50 — The prototype is currently only working in the region of Japan. However, the goal is to create these models in all the big quake zones around the world.
