00:02 Exploring the significance of neuromorphic computing in neuroscience 02:22 Neuromorphic technology offers energy-efficient and fast computing 04:38 Neuromorphic computing mimics the brain's efficiency and uses probabilistic computing. 06:53 Neural network operation - forward and back propagation 08:58 Neuromorphic computing advancements and energy efficiency 11:14 Biological brains have superior computational capabilities over silicone devices. 13:27 Neuromorphic processors enable efficient on-orbit data processing and storage. 15:44 Neuromorphic computing market is rapidly growing. 17:42 Neuromorphic computing is a unique type of computer technology.
Great Video, nice to see Neuromorphic getting some attention. My only complaint is that it is great to just think of NM as low powered DL. There are a lot more differences between DL and NM, NM aims to sense and processing in a sparse and asynchronous manner. It is also spike-based so backpropagation isn't just a given in Spiking Neural Network, so there are many different maybes to allowing backpropagation to work, combine with the numerous method of Hebbian learning that is often used within NM system that helps to further the advantages of low power and latency system.
Thank you for this excellent explanation, it really helped to see what can be done in the future. Neoromorphic computing will clearly grow and I believe it will merge with the classical one in one point for the smoother transition. Still the most awesome is still the biological brain, however it deteriorates and degrades over time and when "trained" or "wired" (from birth) incorrectly in a wrong way it can cause more harm than help. In other words, the gits we have like our brain is our worst and most dangerous enemy!
Great insight 🔥☑️-Neuromorphic computing, AI, ML, and DL are distinctyet interconnected fields. Neuromorphic computing provides brain-inspired hardware and principles that can enhance the efficiency and adaptability of AI systems. AI encompasses a broad range of techniques aimed at creating intelligent behavior, while ML and DL focus on learning from data, with DL employing deep neural networks to handle complex tasks. Understanding these differences and their synergies is crucial for advancing the field of artificial intelligence and developing innovative, efficient computational systems!..
I will prefer some chemical processor with fiber optic data and signal bus that too in parallel to every modular chemical processor or core in unit in future instead of silicone or fiber optics only.
Here is an autonomous driving problem: AI automobile comes upon a situation in which it must choose between running over an old lady who suddenly starts to cross the road, or swerving and hitting a car in which it is virtually certain that a little baby in the rear seat will be severely injured or killed. What does it do? How can that decision be taught using probabilistic or neuromorphic computing? If one or the other were killed, who would be sued for damages? What would you do if _you_ were driving? Just mull that one over for while... It is possible for this situation to occur... And that seems to me to be sufficient motivation for absolute prohibition of autonomous driving. Just forget it.
How did you measure human brain FLOPS? I can do at most one floating point number operation in my mind. And if you give me a big one, it might take me a good dozen of seconds.
NEAT , what if we were able to create the neuromorphic processor on the atomic scale like hydrogen atoms. I think that would make dar.. pa alot of sense. I'm just imagining some crazy mad scientist sitting in his garage when he creates this breakthrough and starts thinking like Dr. Octopus. "Hum, now how could I take over the world with this... or at least get Cara Delevingne to marry me." I've heard of hydrogen atoms being used as memory much like the nodes on a neural network where atoms and their orientation hold differences in more than binary quantifications and I heard DARPA has a pretty NEAT program on reading precognitive thought. I think there are technologies out there most of us are not yet privy to due to the inconvenient nature of their displacement of societies moral norms.
Damn wrong! Mixing ANN and SNN is not a big deal, you explain neuromorphic computing with an ANN?? Why?? Neuromorphing computing actually use SPIKES and not activation functions
Understand all holy books which says same thing about freedom from this perishable body where there is no death just immortality Michio kaku in a conference in Russia told that first solve the problem of just creating another copy of conscious to be able to transfer the conscious into the metal body and thus becoming celestial beings to leave Earth and explore the universe 🤖 Because Carl Sagan once said there are many things in space which might collide with earth Credits: Michio Kaku / Carl Sagan neurallink; Elon musk neuromorphic chips: intel
Good video! And we are capable of reproducing that computational power multiple times as our offspring and provide programming in our homes, schools, and churches that cultivates an enviroment of love, peace, and happiness. All glory to He who is our designer!
What a lot of waffle, probably half the processors on the planet are based on what's called the Harvard architecture. In fact even ARM now uses Harvard since ARM9. I get the vibe that he knows all the words but not what they actually mean.
Your somewhere on planet Earth? So you know the Bug 🪲🐛 lady? Collections of your bug selectives? It's asic Stuff for science 🧪? We're does homework starts at? Lol
I can’t believe you actually mentioned Brainchip Akida! Well done
Thank 🙇
You should do a deep dive into Brainchips Akida Neuromorphic processor. I truly believe that this is an absolute game changer in this space.
Thank 🙇, I will check that
@@ScienceScout Definitely worth your while checking out mate 👍
Brainchip in production, first mover advantage
O it's what planet Earth adds to the extra sector's at space 🌌🚀
Very informative. I like the graphics.
Thank you very much
Transformers like robotic beings with photonic neuromorphic chips
👍
So will done very precise explained
Impressive information 👍🏽👍🏽👍🏽
Soooo good i love his work
00:02 Exploring the significance of neuromorphic computing in neuroscience
02:22 Neuromorphic technology offers energy-efficient and fast computing
04:38 Neuromorphic computing mimics the brain's efficiency and uses probabilistic computing.
06:53 Neural network operation - forward and back propagation
08:58 Neuromorphic computing advancements and energy efficiency
11:14 Biological brains have superior computational capabilities over silicone devices.
13:27 Neuromorphic processors enable efficient on-orbit data processing and storage.
15:44 Neuromorphic computing market is rapidly growing.
17:42 Neuromorphic computing is a unique type of computer technology.
Great Video, nice to see Neuromorphic getting some attention. My only complaint is that it is great to just think of NM as low powered DL. There are a lot more differences between DL and NM, NM aims to sense and processing in a sparse and asynchronous manner. It is also spike-based so backpropagation isn't just a given in Spiking Neural Network, so there are many different maybes to allowing backpropagation to work, combine with the numerous method of Hebbian learning that is often used within NM system that helps to further the advantages of low power and latency system.
Thank you very much 😊
Wow in depth explanation , can u make a video about photonic and quantum computers
I will try to find something interesting about it
@@ScienceScout oh u replied,thanks for replying
Thank you for this excellent explanation, it really helped to see what can be done in the future. Neoromorphic computing will clearly grow and I believe it will merge with the classical one in one point for the smoother transition. Still the most awesome is still the biological brain, however it deteriorates and degrades over time and when "trained" or "wired" (from birth) incorrectly in a wrong way it can cause more harm than help. In other words, the gits we have like our brain is our worst and most dangerous enemy!
Great insight 🔥☑️-Neuromorphic computing, AI, ML, and DL are distinctyet interconnected fields. Neuromorphic computing provides brain-inspired hardware and principles that can enhance the efficiency and adaptability of AI systems.
AI encompasses a broad range of techniques aimed at creating intelligent behavior, while ML and DL focus on learning from data, with DL employing deep neural networks to handle complex tasks. Understanding these differences and their synergies is crucial for advancing the field of artificial intelligence and developing innovative, efficient computational systems!..
Yes sir 🔥🔥🔥🔥
I love this. 😍😍
The imagery is amazing.💯💯
Thank 🙇
Thank 🙇
Thank 🙇
10:47 "PEDDLE FLOPS" lmao
Hahaha!
CMOS is actually pronounced sea-moss
I will prefer some chemical processor with fiber optic data and signal bus that too in parallel to every modular chemical processor or core in unit in future instead of silicone or fiber optics only.
Here is an autonomous driving problem: AI automobile comes upon a situation in which it must choose between running over an old lady who suddenly starts to cross the road, or swerving and hitting a car in which it is virtually certain that a little baby in the rear seat will be severely injured or killed. What does it do? How can that decision be taught using probabilistic or neuromorphic computing? If one or the other were killed, who would be sued for damages? What would you do if _you_ were driving? Just mull that one over for while... It is possible for this situation to occur... And that seems to me to be sufficient motivation for absolute prohibition of autonomous driving. Just forget it.
AI still cant mimic brains 100% yet. It wont ever till broken chemical technology and biology comes to help hardware industry.
How did you measure human brain FLOPS? I can do at most one floating point number operation in my mind. And if you give me a big one, it might take me a good dozen of seconds.
Bigggg love
Thank you 😊
NEAT , what if we were able to create the neuromorphic processor on the atomic scale like hydrogen atoms. I think that would make dar.. pa alot of sense. I'm just imagining some crazy mad scientist sitting in his garage when he creates this breakthrough and starts thinking like Dr. Octopus. "Hum, now how could I take over the world with this... or at least get Cara Delevingne to marry me." I've heard of hydrogen atoms being used as memory much like the nodes on a neural network where atoms and their orientation hold differences in more than binary quantifications and I heard DARPA has a pretty NEAT program on reading precognitive thought. I think there are technologies out there most of us are not yet privy to due to the inconvenient nature of their displacement of societies moral norms.
❤
It's pronounced Noyman.
Akida 1000 💪🏼
Peddle Flops = peta flops?
An exaflop computer still could replace the brain
Is it sentient?
Sentience cannot be measured so how tf are they supposed to know?
It's SILICON, not SILICONE!
Dammit!
Damn wrong! Mixing ANN and SNN is not a big deal, you explain neuromorphic computing with an ANN?? Why?? Neuromorphing computing actually use SPIKES and not activation functions
Understand all holy books which says same thing about freedom from this perishable body where there is no death just immortality
Michio kaku in a conference in Russia told that first solve the problem of just creating another copy of conscious to be able to transfer the conscious into the metal body and thus becoming celestial beings to leave Earth and explore the universe 🤖
Because Carl Sagan once said there are many things in space which might collide with earth
Credits: Michio Kaku / Carl Sagan
neurallink; Elon musk
neuromorphic chips: intel
Tic tac toe!!!!!!????? Reeeeeaaaaallly?
Good video!
And we are capable of reproducing that computational power multiple times as our offspring and provide programming in our homes, schools, and churches that cultivates an enviroment of love, peace, and happiness.
All glory to He who is our designer!
What a lot of waffle, probably half the processors on the planet are based on what's called the Harvard architecture. In fact even ARM now uses Harvard since ARM9. I get the vibe that he knows all the words but not what they actually mean.
Your somewhere on planet Earth? So you know the Bug 🪲🐛 lady? Collections of your bug selectives? It's asic Stuff for science 🧪? We're does homework starts at? Lol
Follow dashingfrog, the big guy in the currency circle will take you to play and make money