"Researchers at Georgia Tech are advancing neural networks to mimic human decision-making by training them to exhibit variability and confidence in their choices, similar to how humans operate, as demonstrated in their study published in Nature Human Behaviour. Their model, RTNet, not only matches human performance in recognizing noisy digits but also applies human-like traits such as confidence and evidence accumulation, enhancing both accuracy and reliability. Credit: SciTechDaily.com" (ScitechDaily, AI Learns To Think Like Humans: A Game-Changer in Machine Learning)
The AI learns to think like a human. That is a breakthrough in machine learning. And those systems are important for controlling robots. The machine, which thinks like a human is the tool, that can control quantum computers. AI-based operating systems can also make morphing neural networks more powerful than before.
The main thing in morphing neural networks is that there is always a gate open when the system wants to drive more data into that system. If there is stuck somewhere in the morphing neural network, the system can drive data using another route.
One of the reasons why it's so hard to model machines that think like humans is that we always think that the brain is a 3D structure. However, researchers can model those structures into the 2D model. In that model, the flat brain's main structure looks like a tricolor. The computer or its database's main structure is divided into three main units, that act like human brains. Each database can have its, own individual physical microprocessor, that controls its action.
The model of the 2D brains is like a tricolor. Each color symbolizes each part of the brain's main structure, cortex, midbrain, and modula oblongata.
The machine that thinks like a human is theoretically very easy to make. The system requires three different main systems, that act like cortex, midbrain, and modula oblongata. The data travels between those structures. When we think about the human brain we can copy all those structures into the substructures.
The idea is simple. We can put the drawings of human brains on the floor and then put microchips on the image. Then we can share a mission with those microchips following that chart. That makes it possible to create a new and powerful database structure.
The artificial neural network can have organic components. That component can be electric eel cells that produce electricity for those robots. Or they can be cloned neurons. And that means things like drone swarms can be as intelligent as humans.
Above: The neural network diagram. Each knot point can be a drone. Or even a quantum computer.
The neural network that is introduced in the second image is one of the most powerful tools in the world. Each of the points in the diagram can be a drone. That makes it possible to create a traveling artificial morphing network, which can solve complex problems.
But artificial neural networks can be more intelligent than we even think. The artificial neural network can involve the organic actor. That organic actor can be the electric eel's cells, that create electricity for those drones. Or the system can involve living neurons, that communicate with microprocessors. Those systems are in lab-grown mini-brains, and those drone swarms or robot groups can be as intelligent as humans.
Those robots require nutrients, and if they have an artificial stomach with the right bacteria, those machines can eat like normal humans. Living cells require nutrients. They also require an immune system, that protects them against outside threats like microbes.
And that is one of the most interesting and frightening technologies in the world. We must realize that when we create intelligent machines, we might someday create an alien or modern Frankenstein. The machine that might make us think about our relationship to technology.
https://scitechdaily.com/ai-learns-to-think-like-humans-a-game-changer-in-machine-learning/
Comments
Post a Comment