From The Brain To AI (What Are Neural Networks)

Visit Our Parent Company EarthOne ➤

In the last video in this series we discussed the differences between deep learning and machine learning, how and when the field of deep learning was officially born, and it’s rise to mainstream popularity. The focus of this video then will be on artificial neural networks, more specifically – their structure.

00:00 Intro
00:27 From The Brain To AI

Thank You To The Members Who Supported This Video ➤

Wyldn Pearson
Garry Ttocsra
Brian Schroeder

Become A Member & Help Us Grow ➤

Learn More About Us Here ➤

Join Our Discord ➤

Soundtrack ➤

♫ 00;00 "Clair de Lune" by RELAYER
♫ 00;27 "Sun" by HOME
♫ 02;53 "Flood" by HOME
♫ 06;09 "Resonance" by HOME
♫ 07;49 "June" by Aire Atlantica

Sources ➤

Producer ➤ Ankur Bargotra

Follow The Producers Social Media Accounts ➤

[VishwanathBondugula]

Next level visualization and knowledge representation

[mrintention]

Singularity Prosperity in the future.🔥💥🙌🔥

[avi]

The amount of information explained in such short video is praiseworthy ...well done

[mrmasj]

Awesome to see an increased amount of videos coming, keep up the good work i love your production quality :)

[tyfoodsforthought]

High quality stuff as usual! Great job

[КирилПетров-ыф]

For the quality of the video you deserve much more views.

[SNAPPERTBK]

This channel literally a gift to Humankind.

[noisulli]

Amazing channel going on a binge watch

[HansDunkelberg]

As it's here illustrated, a circle representing the cell body of a neuron only can serve as a tiny cogwheel within a larger network. A node processing letters like shown from 5:59 cannot recognize what letters fire into it, it only can recognize that _some_ letters do so. If it fires on because it has received input from A, C, and E, its output will be the same like when it receives an equally strong input from A, B, and / or D. So such a node is just a variable fed from an OR-gate.

Essential for a mimicking of the brain will be the capacity of _establishing new variables, _ especially of establishing new variables which also are dependent on AND-circuits. For example you can make five nodes, each of which only will be triggered by _one_ of those five letters. Then you can make a sixth node which only fires if it has received an input from all of those five nodes (for A, B, C, D, and E). This would mean an intelligent node inasmuch as it recognizes a pattern - here the one of an even distribution of input, for the concept of "all".

I assume such operations to be possible in the brain because the signals processed in the brain _fade, _ over time. Such a temporal restriction will be necessary because you sooner or later will need a neuron for other tasks, and if you then still have old input stored by it, you'll obtain a distorted reaction.

How might our brains be able to establish new variables? Obviously they can do it, given how many abilities we can learn which we do not possess at our births. Does the brain again and again grow new axons, for an establishment of such novel connections, or what other means might it have to enable such a reprogrammability?

It would interest me likewise how the brain organizes its obviously existing more complicated IF-THEN gates. To recognize the pattern of "none" rather than of "all", for example, you'll certainly need such a tool. The node shown from 5:59 is an IF-THEN gate inasmuch as it only fires _if_ it is filled up itself quickly enough, but what can you do to obtain an output which expresses the _lack_ of a certain input?

[Patrick-wnuj]

this is the first time i have understood what a neural network is

[CuzzaFuturism]

A very informative video. I appreciate the intricacy of the editing, it's incredible! You deserve more views. 👍

[DerekFolan]

It's amazing how the structure can lead to such processing for understanding, these structures are in our own heads and we dont even know how they work, Marvellous though, these weird abstract structures let us get an answer to something. Maybe general learning is combining all the answers from the first analysis and doing a deeper analysis. Loops within loops. But what is the best structure. Well like people it probably turns out there is not a best structure, the final output is different forms of intelligence, varied, like all the colours of the rainbow, just like people. Wonderful when you gaze up at the stars and imagine how big the universe is, and then you realize connections in the brain are nearly as massive

[SakshiVats-gjxh]

How to disconnected for mind neural networks to other devices

[vinniepeterss]

It's a shame this channel don't upload more😢

[UBIAICOSMOS]

Artificial Intelligence is definitely the future of the world.
AI will drive the economy of tomorrow
Universal Basic Income - dignity Life For every one.
Basic Income is a human right
Technology is cheaper than workers.Time for #BasicIncome, Four day work week, 3days off weekly.

[evilovesperry]

Does anyone else not see that naruto taught all of this with shadow clones