The researchers explained that the models effectively remember any information. This is usually fine, but in
This method is called Expire-Span and will helpneural networks efficiently sort and store information that is most suitable for performing the tasks assigned to them. Expire-Span works by first predicting what information will be most useful to the network in a given context, and then assigning an expiration date to it. The more important, according to Expire-Span, this or that information, the more distant the shelf life of the data. Neural networks will be able to keep relevant information for a longer time, constantly freeing up memory, forgetting irrelevant data.
Created a robot that combes even the most tangled hair
Expire-Span calculates the expiration dateinformation for all kinds of data and does this every time he is presented with a new piece of information. This tag determines how long information is retained in memory, the researchers explained. "The gradual decay of a piece of data is the key to preserving important information without blurring it."
Although research is still in its early stages,Scientists are confident that not many steps are needed to improve the neural network. In the future, the team hopes to develop an even closer human memory, but capable of assimilating new information much faster than current technology allows.
Scientists have come up with a way to charge electric vehicles on the go
Scientists have shown how a black hole tears apart a star
Evolutionary error: which organs in the human body work illogically