Welcome to DU!
The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards.
Join the community:
Create a free account
Support DU (and get rid of ads!):
Become a Star Member
Latest Breaking News
General Discussion
The DU Lounge
All Forums
Issue Forums
Culture Forums
Alliance Forums
Region Forums
Support Forums
Help & Search
Neuromorphic Nanowire Networks learn and retain memory like human synapses.
Scientists have created a new type of artificial brain that can learn and remember things on the go, just like our brains do. This could lead to more innovative and greener machines that can handle complex and changing data in the real world.
The artificial brain is made of nanowire networks, tiny wires a billion times smaller than a meter. The cables form random patterns that look like the game ‘Pick Up Sticks,’ but they also act like the neural networks in our brains. These networks can process information in different ways.
The nanowire networks can learn and remember using simple rules that change the electrical resistance at the points where the wires cross. This is called ‘resistive memory switching,’ which happens when the cables get more or less conductive depending on the electrical signals they receive. This is similar to how our brain cells, or neurons, communicate with each other through synapses.
She said that their previous work showed that nanowire networks could remember simple tasks, but this work showed that they could also handle dynamic data that came online. She said this was a significant advance because online learning was hard to do with large and changing data sets. The usual way to do machine learning was to store the data first and then train a model on it.
The artificial brain is made of nanowire networks, tiny wires a billion times smaller than a meter. The cables form random patterns that look like the game ‘Pick Up Sticks,’ but they also act like the neural networks in our brains. These networks can process information in different ways.
The nanowire networks can learn and remember using simple rules that change the electrical resistance at the points where the wires cross. This is called ‘resistive memory switching,’ which happens when the cables get more or less conductive depending on the electrical signals they receive. This is similar to how our brain cells, or neurons, communicate with each other through synapses.
She said that their previous work showed that nanowire networks could remember simple tasks, but this work showed that they could also handle dynamic data that came online. She said this was a significant advance because online learning was hard to do with large and changing data sets. The usual way to do machine learning was to store the data first and then train a model on it.
More here: https://interestingengineering.com/innovation/artificial-brain-learns-on-the-fly-with-nanowire-networks?
___________________________________________________________
The singularity is coming at us hard and fast.
2 replies
= new reply since forum marked as read
= new reply since forum marked as read
