Showing posts with label Brain. Show all posts
Showing posts with label Brain. Show all posts

Tuesday, September 5, 2023

Bees are making decisions more effectively than humans.

  Bees are making decisions more effectively than humans.


The reason why researchers are researching the bees' nervous system is simple. Researchers can use this kind of data can use to develop next-generation AI-based infrastructures. The CPU in non-centralized neural networks only delivers missions to the system. 

The idea is that the CPU (Central Processing Unit) delivers the mission to the sub-system. That can be the RISC (Reduced instruction set computer system). When we think about the human-looking robots, the pneumatic-hydraulic system and the microprocessors that control them are RISC systems. 

Many times before this researchers claim that insects make faster and more precise than humans. There are a couple of reasons why insects like bees make better decisions than humans. If we compare the bees to humans bees are like RISC operators. RISC-operator is a computer that has fewer actions than a regular PC. 

That makes RISC -devices more effective than PCs but their operational sector or skills are more limited than regular PCs have. If we compare a bee with a human that situation is similar as we compare a pocket calculator with a PC. We can make basic calculations faster if we use a pocket calculator. 

Then if we use a PC for that thing we need a computer program and the regular keyboard is not as fast as a pocket calculator. But a regular PC can do many other things. Then just calculate some basic calculations. The number of skills of that computer depends on the data and programs that it has access to. 

The bee has a limited number of skills. And that means there are not many choices that the bee's nervous system can use to select the necessary operations. In the bee's nervous system are not many neurons. And that means the bee must not go through about 100 billion neurons when the bee wants to make something. There are about 950,000 neurons in the bee's brain.

In the brain, the number of physical neurons is not important in the case of decisions. In the thinking process, the most important thing is how to interconnect memory cells into new virtual neurons that can interconnect their data together. 




"New research unveils the decision-making pathways in bee brains, shedding light on their ability to quickly and accurately assess flowers for nectar, which could inspire more autonomous robot designs. The study, led by various academic experts, also emphasizes the efficiency of evolutionarily refined insect brains that could guide future AI development in industries. Credit: Théotime Colin" (Researchers Discover That Bees Can Make Decisions Better and Faster Than We Do)


In reflexes, the position of those memory cells is more important. The thing that makes the reflex effective is the position of memory cells. If memory cells are close to muscle cells each of those cells includes multiple images. That thing makes them more effective. In this case, the net eyes of the bee send images to neurons. 

The image always activates neurons and it sends signals to muscles. The reason for multiple connections in a bee's brain is that the eyes send images to the memory cells. The thing that determines do the bee escapes or not is how many memory cells activate. So the bee's nervous system doesn't send anything back. 

All impulses travel in the same direction. And when they activate the neurons that control muscles, those cells send the signals to muscles anyway. That means some decisions that bees make are a series of reflexes. That causes the question: Do human thinking and human actions also series of reflexes? What if a bee has 200 billion reflexes?

But I don't know if that number is made by calculating physical neurons or if that number is obtained by calculating virtual neurons, which form when neurons interconnect with other neurons. There are more synaptic connections in bee's brains than in human brains. And my opinion is that. We should rather calculate virtual neurons than physical neurons. 

Every connection between the bee's brain cells same way as in the human brain gives a new state to that system. Theoretically is possible that researchers create a single neuron system that is as intelligent as humans. But that thing requires the neuron to have 300 billion circular connections where the axon is connected to that neuron itself. 

Bee's brains are acting in a similar way to human brains. There are memory cells. And every memory cell involves some sensorial data or reaction for those senses. Some other memory cells tell neurons and the router cells how they must send signals to the muscle cells. 

If we think that a bee has a limited number of actions. We can understand why it's so effective. We can think that all skills are memories. And memories are stored in some kind of warehouse. If the number of skills that bees have is like a book, the human has warehouse class skills and abilities. If we had only about 5 or four skills and we had to select from them the selection process would be very fast. But if we have 100 or more skills that gives a response to the event. 

That our senses send to us that thing takes far longer time than in the case that we have only 5 or fewer skills. All skills that humans have formed from sub-skills. So the network or the puzzle of neurons that form virtual neurons is making all our skills. Our skills are more complicated than some bees whose only mission is to find food and then return to the nest. So the only skill that the bee needs is how to find food. And then how to avoid the predators. 


https://scitechdaily.com/researchers-discover-that-bees-can-make-decisions-better-and-faster-than-we-do/

https://en.wikipedia.org/wiki/Reduced_instruction_set_computer



Tuesday, August 22, 2023

Forgetting is one of the most important things in the brain.

Forgetting is one of the most important things in the brain.


Forgetting is as important as learning.


Forgetting things is necessary for people. If people are not forgetting things, they will have a very large memory. And large memory storage makes the brain slow. The reason brains forget is that they can respond to things that require a fast reaction. This is the reason why people have three memory levels.

How often some neural connection is needed determines the level of memory. When the brain does not use some neural connections, it removes some of them. That transfers memory backward. If the neural connection is not used or needed, the brain removes that connection, releasing memory cells for some other purpose.

In the human brain, there are 100 billion to 300 billion neurons. Depending on how we calculate them. If we calculate only individual neurons, there are 100 billion neurons. But if we calculate virtual neurons, which means the neuron combinations that axons connect, the number of neurons can be 300 billion. All memories form when axons connect neurons.



"Neuroscientists suggest that “forgetting” might be a form of learning that benefits flexible behavior in dynamic environments, and their experimental tests demonstrate that memories are not truly lost but are stored in engrams that can be reactivated. Their findings, which have implications for conditions like Alzheimer’s disease, show that memories can be retrieved through both natural and artificial cues." (ScitechDaily.com/“Forgetting” Might Not Be a Bad Thing – Scientists Propose That It Could Be a Functional Feature of the Brain)


The purpose of memory is to ensure survival. Memory is not meant for learning things like mathematical formulas. It's made to remember dangerous things. This is why our memory stores negative things more strongly than positive ones. And this is the reason why negative things remain longer in memory than positive things.

When brains require some memories, they must send queries to neurons to see if they have the memories that a person needs stored in their structure. Every skill we have is stored in memory. Lots of memories mean that the brain must send more queries than the brain with fewer memories.


Why is forgetting a good ability for learning machines as well?


Researchers use brains and their functions for modeling machine learning. Machine learning means that the computer stores solutions that it makes by following certain parameters. The problem with learning machines is that they operate in complicated environments. They must collect lots of data.

And when those systems require a certain solution, they must check the entire database structure for all records one by one to find a solution. That matches certain situations. To make databases simpler and faster. The machine must remove data that it doesn't need. Without that ability, databases will grow too complicated and too big to give fast responses to situations that machines will face.


https://scitechdaily.com/forgetting-might-not-be-a-bad-thing-scientists-propose-that-it-could-be-a-functional-feature-of-the-brain/ 


https://technologyandfuture4.wordpress.com/2023/08/22/forgetting-is-one-of-the-most-important-things-in-the-brain/

Could fast-spinning aerial vehicles be behind some UAP cases?

“The Phantom Twist drone’s unique rotation renders it almost invisible when in flight (Image Credit: Northwestern University).” (The Brief) ...