Long Short-Term Memory

stock here.  

Amazing a high level government troll who used to troll ENENEWS, and posts here as Loose Nuke, has returned, but with something of value.

LSTM networks are well-suited to classifying, processing and making predictions based on time series data, since there can be lags of unknown duration between important events in a time series. LSTMs were developed to deal with the vanishing gradient problem that can be encountered when training traditional RNNs.

It immediately occurred to me that this is exactly a problem in Earthquake prediction, as the "gaps" have a huge standard deviation, and the unusual is THE USUAL.

I propose to combine the LSTM with my proprietary numerical approach to solving systems.    Equations that might take days to develop can often be developed in less than an hour by using "numerical methods" which is a combination of math, statistics, observation, and intuition, followed by experimentation to confirm or break "the system".

More to come on this.  

---------------------------------------------------
Much of geo-sciences have been "accepted" for decades.   Methinks much of it will be flipped on it's head in the next few years.   Here is one example.

https://beforeitsnews.com/science-and-technology/2020/05/ancient-rocks-show-high-oxygen-levels-on-earth-2-billion-years-ago-2976487.html