Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

📁 Terence Tao, Fields Medalist mathematician, explains that neural networks do not understand concepts, but they uncover hidden patterns. In knot theory, a classical neural network discovered correlations between mathematical invariants no one had anticipated. It started as a black box, then became a real clue to deep structures... show more

Jon Hernandez

23,666 subscribers

204,209 views • 6 months ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Ever wondered what neural networks are and how they work? Systems like ChatGPT use neural networks to work as well as they do. Neural networks are composed of "layers" of neurons, layers with different functions; connections between layers called "weights"; and mathematical functions called "activation functions". If you’re interested in learning about these systems, check the comments. Ultimately, the neural network structure of the model serves to visually demonstrate that it is, in fact, a complex mathematical equation. When companies release the model's weights, they are releasing a key component needed to run the model's complete equation. Without the weights, the equation is incomplete. For the math-minded: the weights of a model are the learned numbers (they are variables during training) that are then used as constants in the mathematical functions that make up the model. Neural networks are ultimately just one big, hyper-complex mathematical function, and when a model is trained, it learns the constants associated with the high-dimensional input.

Ever wondered what neural networks are and how they work? Systems like ChatGPT use neural networks to work as well as they do. Neural networks are composed of "layers" of neurons, layers with different functions; connections between layers called "weights"; and mathematical functions called "activation functions". If you’re interested in learning about these systems, check the comments. Ultimately, the neural network structure of the model serves to visually demonstrate that it is, in fact, a complex mathematical equation. When companies release the model's weights, they are releasing a key component needed to run the model's complete equation. Without the weights, the equation is incomplete. For the math-minded: the weights of a model are the learned numbers (they are variables during training) that are then used as constants in the mathematical functions that make up the model. Neural networks are ultimately just one big, hyper-complex mathematical function, and when a model is trained, it learns the constants associated with the high-dimensional input.

Harper Carroll

29,098 views • 8 months ago

📁 Geoffrey Hinton, a pioneer of deep learning, says LLMs do understand what they say. He explains that understanding language is not translating it into an internal symbolic code. In neural networks, words act like Lego blocks that combine to form meaning. It is not translation, it is structure. That is what real language understanding looks like.

📁 Geoffrey Hinton, a pioneer of deep learning, says LLMs do understand what they say. He explains that understanding language is not translating it into an internal symbolic code. In neural networks, words act like Lego blocks that combine to form meaning. It is not translation, it is structure. That is what real language understanding looks like.

Jon Hernandez

77,663 views • 5 months ago

📁 Terence Tao, mathematician and Fields Medalist, says AI may look like it reasons, but it does not understand. Language models imitate the steps of human reasoning without solid foundations. Sometimes they get it right, sometimes they collapse entirely. The output looks convincing, but it is fragile. It feels like understanding, yet it is probability pretending to think.

📁 Terence Tao, mathematician and Fields Medalist, says AI may look like it reasons, but it does not understand. Language models imitate the steps of human reasoning without solid foundations. Sometimes they get it right, sometimes they collapse entirely. The output looks convincing, but it is fragile. It feels like understanding, yet it is probability pretending to think.

Jon Hernandez

252,332 views • 6 months ago

Hopfield recall Spiking Neural network now this one is cool isn't it? -open source, link in comment- This is a Spiking Neural Network (SNN) that can learn and recall patterns through Hebbian learning, similar to a Hopfield network but using biologically-inspired spiking dynamics rather than energy-based settling.

Hopfield recall Spiking Neural network now this one is cool isn't it? -open source, link in comment- This is a Spiking Neural Network (SNN) that can learn and recall patterns through Hebbian learning, similar to a Hopfield network but using biologically-inspired spiking dynamics rather than energy-based settling.

echo.hive

31,452 views • 5 months ago

Terence Tao says the era of AI is proving that our definition of intelligence is inaccurate We thought intelligence was some vague, mystical way of thinking But as AI solves tasks, it never looks intelligent, just tricks, neural networks, and next-token prediction "maybe that's actually a lot of what humans do"

Terence Tao says the era of AI is proving that our definition of intelligence is inaccurate We thought intelligence was some vague, mystical way of thinking But as AI solves tasks, it never looks intelligent, just tricks, neural networks, and next-token prediction "maybe that's actually a lot of what humans do"

Haider.

484,704 views • 5 months ago

This project argues that latent space—the learned, multidimensional, representational substrate of neural networks—is not only a mathematical construct but, read through literature, a structure akin to architectural interiority that renders the abstract physically legible.

This project argues that latent space—the learned, multidimensional, representational substrate of neural networks—is not only a mathematical construct but, read through literature, a structure akin to architectural interiority that renders the abstract physically legible.

Gašper Beguš

10,671 views • 3 months ago

I animated how a neural network learns, from one neuron to backpropagation, simple math, no calculus required. Best 8 minutes you can spend if it has only ever been a black box to you.

I animated how a neural network learns, from one neuron to backpropagation, simple math, no calculus required. Best 8 minutes you can spend if it has only ever been a black box to you.

tetsuo

46,251 views • 10 days ago

Does GPT understand the world? Here is what Ilya Sutskever, co-founder of OpenAI, says during a discussion with Jensen Huang, CEO of Nvidia: (1) When we train a large neural network to accurately predict the next word in lots of different texts from the internet, the AI is learning a world model. (2) On the surface, it may look like learning correlations in text, but it turns out that to 'just learn' statistical correlations in text, to compress information really well, what the neural network learns is some representation of the process that produced the text. (3) This text is a projection of the world...what the neural network is learning is aspects of the world, of people, of the human conditions, their hopes, dreams, motivations, their interactions...the situations we are in. The neural network learns a compressed, abstract, usable representation." Do you think learning representations = understanding? Are large language models simply stochastic parrots, or are they much more?

Does GPT understand the world? Here is what Ilya Sutskever, co-founder of OpenAI, says during a discussion with Jensen Huang, CEO of Nvidia: (1) When we train a large neural network to accurately predict the next word in lots of different texts from the internet, the AI is learning a world model. (2) On the surface, it may look like learning correlations in text, but it turns out that to 'just learn' statistical correlations in text, to compress information really well, what the neural network learns is some representation of the process that produced the text. (3) This text is a projection of the world...what the neural network is learning is aspects of the world, of people, of the human conditions, their hopes, dreams, motivations, their interactions...the situations we are in. The neural network learns a compressed, abstract, usable representation." Do you think learning representations = understanding? Are large language models simply stochastic parrots, or are they much more?

Alex Ker 🔭

1,367,023 views • 2 years ago

This is a comparison video between LI and Tesla's autonomous driving systems. LI once claimed in 2024 that it had developed an end-to-end neural network better than Tesla's, but what the truth is, we can see it in this video.

This is a comparison video between LI and Tesla's autonomous driving systems. LI once claimed in 2024 that it had developed an end-to-end neural network better than Tesla's, but what the truth is, we can see it in this video.

𝐶ℎ𝑎𝑟𝑙𝑒𝑠

35,248 views • 1 year ago

Terence Tao: "It's HUMBLING to be a Mathematician"

Terence Tao: "It's HUMBLING to be a Mathematician"

Prof. Brian Keating

21,660 views • 6 months ago

samcedes had real chemistry, real tension, real emotion… but glee had no clue how to write a couple that actually worked

samcedes had real chemistry, real tension, real emotion… but glee had no clue how to write a couple that actually worked

vini ミ☆

199,419 views • 10 months ago

Individual Neuron: Neural Network A neuron in a neural network performs a weighted sum of inputs, adds a bias, and applies an activation function like sigmoid, ReLU, or tanh, introducing non-linearity. This output helps the neuron learn and represent patterns in the data.

Individual Neuron: Neural Network A neuron in a neural network performs a weighted sum of inputs, adds a bias, and applies an activation function like sigmoid, ReLU, or tanh, introducing non-linearity. This output helps the neuron learn and represent patterns in the data.

ₕₐₘₚₜₒₙ — e/acc

47,467 views • 2 years ago

Motion planning in complex tasks is hard and still done via slow, explicit, traditional planners. We present a generalist Neural Motion Planner -- a single neural network that plans complex dynamic motions quickly and accurately at test time. Building upon our lab's sim2real efforts, the key idea is to create many complex scenes in simulation and then distill classical motion planner trajectories into a single reactive neural network policy. More details in the thread below! 👇 Open-sourced:

Motion planning in complex tasks is hard and still done via slow, explicit, traditional planners. We present a generalist Neural Motion Planner -- a single neural network that plans complex dynamic motions quickly and accurately at test time. Building upon our lab's sim2real efforts, the key idea is to create many complex scenes in simulation and then distill classical motion planner trajectories into a single reactive neural network policy. More details in the thread below! 👇 Open-sourced:

Deepak Pathak

28,642 views • 1 year ago

Fools call it a "dead language"....... Meanwhile, we are debating the nuances and applications of AI, neural networks, and the future of artificial intelligence on national television (DD News) exclusively in classical Samskrutam (Sanskrit). A language engineered with mathematical precision is the perfect medium to discuss computation.

Fools call it a "dead language"....... Meanwhile, we are debating the nuances and applications of AI, neural networks, and the future of artificial intelligence on national television (DD News) exclusively in classical Samskrutam (Sanskrit). A language engineered with mathematical precision is the perfect medium to discuss computation.

prathosh ap

12,751 views • 2 months ago

A phone, a house, and some neural networks

A phone, a house, and some neural networks

Luma

138,297 views • 3 years ago

JANE STREET PAYS $650,000 A YEAR FOR QUANTS WHO TRADE USING NEURAL NETWORKS. STANFORD JUST DROPPED THE COMPLETE NEURAL NETWORK FUNDAMENTALS LECTURE FOR FREE. BOOKMARK & WATCH THIS TODAY, THEN READ HOW QUANTS APPLY THIS IN THE ARTICLE BELOW BEFORE SOMEONE TAKES IT DOWN.

JANE STREET PAYS $650,000 A YEAR FOR QUANTS WHO TRADE USING NEURAL NETWORKS. STANFORD JUST DROPPED THE COMPLETE NEURAL NETWORK FUNDAMENTALS LECTURE FOR FREE. BOOKMARK & WATCH THIS TODAY, THEN READ HOW QUANTS APPLY THIS IN THE ARTICLE BELOW BEFORE SOMEONE TAKES IT DOWN.

Roan

126,223 views • 2 months ago

The universal approximation theorem states that a neural network with one hidden layer can approximate continuous functions on compact sets with any desired precision.

The universal approximation theorem states that a neural network with one hidden layer can approximate continuous functions on compact sets with any desired precision.

ₕₐₘₚₜₒₙ

218,755 views • 2 years ago