yego.me
💡 Stop wasting time. Read Youtube instead of watch. Download Chrome Extension

How AI, Like ChatGPT, *Really* Learns


2m read
·Nov 7, 2024

The main video is talking about a genetic breeding model of how to make machines learn. This method is simpler to explain or just show. Here is a machine learning to walk, or play Mario, or jump really high. A genetic code is an older code, but it still checks out, and I personally suspect in the future genetic models will have a resurgence as compute power approaches crazy pants.

However, the current hotness is deep learning and recursive neural networks, and that is where the linear algebra really increases and explainability in a brief video really decreases. But if I had to kind of explain how they work in a footnote, just for the record, it's like this: No infinite warehouse. Just one student. Teacher Bot has the same test, but this time Builder Bot is 'Dial Adjustment Bot,' where each dial is how sensitive one connection in the student bot's head is.

There's a lot of connections in its head, so a lot of dials. A LOT, a lot. Teacher Bot shows Student Bot a photo, and Dial Adjustment Bot adjusts that dial stronger or weaker to get Student Bot closer to the answer. It's a bit like adjusting the dial on a radio. Is that still a thing? Do cars have radios still? I don't know, anyway.

You might not know the exact frequency of the station, but you can tell if you're getting closer or further away. It's like that but with a hundred thousand dials and a lot of math, and that's just for one test question. When Teacher Bot introduces the next photo, Dial Adjustment Bot needs to adjust all the dials so that Student Bot can answer both questions. As the test gets longer, this becomes an insane amount of math and fine-tuning for Dial Adjustment Bot.

But when it's done, there's a student bot who can do a pretty good job at recognizing new photos, though still suffers from some of the problems mentioned in the main video. Anyway, that's the most babies' first introduction to neural networks you will ever hear. If it sounds interesting to you and you like math and code, go dig into the details; machines that learn are the future of everything.

Maybe, quite literally, the future of everything, and given what we've put them through, may the bots have mercy on us all.

More Articles

View All
Ray Dalio: STOP Trying to Time the Market
Wouldn’t it be great if we as investors could know right when the stock market was about to fall so we could get out and not lose money? And then jump right back in at the bottom of the market? Based on some of the videos on YouTube and articles written o…
In the Same Boat | Port Protection
Yeah, let’s put the bow right up against her. Port protection main state Timothy Curly leech and newest resident Amanda Ma are getting Curly’s vessel, the little Pelican, ready to go hbit fishing. “My boat is one of the most important things that I own. …
Derivatives expressed as limits | Advanced derivatives | AP Calculus BC | Khan Academy
Let’s see if we can find the limit as h approaches 0 of (5 \log(2 + h) - 5 \log(2)), all of that over (h). And I’ll give you a little bit of a hint, because I know you’re about to pause the video and try to work through it. Think of your derivative proper…
Khanmigo Teacher Story - Ms. Bartsch
What I love most about using Conmigo is it gives me the agency as a teacher to be able to kind of set parameters for my classroom while still giving my students exposure to the AI that’s going to be a huge part of the world that they’re heading towards af…
Big takeaways from the Civil War
We’ve been discussing the American Civil War, which lasted from 1861 until 1865. It was the deadliest conflict in all of American history, in which about 620,000 Americans lost their lives. We briefly went over the very end of the war, as Grant caught up …
Filming Cliff-Jumping Geese: On Location | Hostile Planet
The animals who are filmed for “Hostile Planet” have to survive in incredibly tough conditions. But they’re adapted to it. The crew, on the other hand, that’s a whole different ball game. RENEE GODFREY: Making a series like “Hostile Planet” wasn’t simple…