Assist me on Patreon: On this video I’ll clarify what the primary variations are between the …
source
Tags: aiArtificialartificial intelligenceartificial intelligence newsartificial intelligence news 2023artificial intelligence vs human intelligencecomputer sciencedifference betweeen human and artificial intelligencedifferencesHumanhuman vs artificial intelligenceIntelligencelatest news about robotics technologylatest robotslatest robots 2023machine learningneural netneural networkrobot newsrobotics newsrobotics news 2023robotics technologies llcrobotics technologyScience
Beautiful.
Daring wardrobe darling.😊
Every day it becomes more and more obvious that AI is the Great Filter, solution to Fermi's Paradox, the next stage of life.
This video is just 3 years old why she looks and sounds so young
There is a bias issue, human brain(s) select the problem, tasks and data sets for neural networks. For example, pre-existing brain structures developed by evolution are optimized for certain functions while neural networks must start from scratch. The full time to train a human brain is actual time needed for species evolution.
Your accent was so exotic back then
grip reaper
damn mommy
I'm not so sure that the neuron is either on or off. This is partially true, with a few caveats, first the rate of firing is variable from one neuron to the next. Secondly, the neuron is continuously weakening or strengthening the connections it has with other neurons via synaptic transformations. The weighting system in artificial neural networks is completely numerical, but it is a way of simulating what actually takes place in the brain, which is a physical reconfiguration of the connections between neurons. A neuron may fire, but the impact of its firing will be greater on those neurons it has stronger connections with and weaker on neurons that it shares weaker connections with. So it does seem like the brain is actually learning in a very similar, but perhaps not identical, way that ANNs do.
Great video….I would be curious to know if it is possible for AI to ever become "smarter" than the person(s) who programmed the logic it uses to make calculations/decisions….if not, that seems that would be a huge limitation on how smart AI can ever become….
In human brain there are countless countless countless countless neurons required to become enlightened.
The weights are free to vary, but the values of the neurons themselves are a result of the weights and input combined. So, I think you're wrong about the actual values of the neurons being free to vary too.
Impressive how much machine learning improved over a few years. Yet final message is still up to date. Some people forget that AI models are just tools and we shouldn't let them make critical decisions just like no one should let a roomba watch over a toddler.
It is an old video, even so, 3 years ago we had already Transformers and Large Language Models, the video seems to refers to Neural Networks from 10 years ago, and not from the moment it was done, or the current moment in time. Not saying it is incorrect, it is a good video overall! but some things (like talking about 300 "aritifial neurons" typically, or that Neural Networks are not able to generalized doesn't seem correct IMHO, almost everybody knows about ChatGPT right now, and I think that it is capable of generalize a lot (maybe not to the level of human yet, but it does it for sure, and it definitely has some sort of World Model, even if it is not 100% correct or consulted "to the letter").
good critique , I have seen where I am working some gang bang professors of math taking grant for AI saying that while AI works the principle of intelligence remain mysterious and they will focus on understading the mathematicl principles of cognitive principles. I was thinking how the hell they could state that bullshit since AI has nothing to do with cognitive intelligence. It is about statistics.
The brain is truly amazing. Let's me know that things in the Universe are 'by design' and not random. What people do with their intellect may be random, but most things in the universe are by design.
I's sorry just destracted by those amazing hips … mmm you look fine, you Germans have some good genetics going for you, I guess you are 40 ish ? And looking that tight ? Jeez ! Oeh look at that, mmm you look fine ! What a body !
As far as I am aware the human brain does not have around 100 billion neurons – it has an average of 86 billion neurons and 85 billion nonneuronal cells.
Now we have Chat GPT 4 with trillions of parameters.
It is not clear now the comparison.
Just a guy on the internet so assume in my Moms basement eating Dorritos BUT '8. . . . no one knows in detail how humans learn but we know thats not how they do it'? That appears to be exactly how humans learn minus the word-play, prediction–>expectation–>measure-deviation–>prediction error correction? We know this is so at least in motor-skill learning. What am I missing, sure we are not doing it in wet-ware but learning itself does not have any other means that I know of? Predict-test-correct-confirm?
Well, I'm gonna downgrade the conversation. You look so sexy in this one.
human neurons also have way more individual behaviour.. not only can the cell body give out signals, each dendrite ("hand of a neuron") can give them out as well, independently.
Thr brain is just a signal translation machine/device for consciousness. Mainstream scientists are too dumb to see the obvious.
What a fine specimen. Beauty, grace, exceptional intelligence, be still my heart.😮
sexy woman