OO this anid oo that. American English please!!

Fantastic story. Thanks for sharing!

I didn't read the cited paper, but if I understood correctly, the 5 billion images need to be labeled for the training step?

Question 🤚 Up until recently all images of a cat on internet were photos of real cars and the system could use them in training. But now more and more cat images are AI generated. If future systems use generated images in training it will be like a blind leading a blind. More and more distortion will be added. Or? Can that be avoided?

The more I learn about AI the more I realise that there is no "intelligence" at all! Seems to be more brute force probability calcs using ridiculous massive computer power. Take the "cat" for example. There is no understanding at all of what a cat is, typical personality of a cat, what a cat needs, how a cat interacts with its environment etc etc. The idea that AI is close to consciousness or self-awareness is complete and utter BS! IMO it's not even close to intelligence

@dr Pound: sorry if this is off topic here but, i wonder if the problem of hallucinations in AI comes from us not treating the 'i don't know what I'm looking at ' answer of a model, as a very negative outcome? If it was treated by us as a valid neutral answer, could it reduce the rate if hallucinations?

Your cat needs a tail!

Can't you do something analogous to stable diffusion for text classification? Get the image embedding, and then start with random noisy text, and iteratively refine it in the direction of the image's embedding to get a progressively more accurate description of the image.

Mike is the best

I thought diffusion models had VAE and not ViT Correct me if I m wrong

I saw a video a while ago where a guy got ChatGPT to explain medical procedures step by step that could be used for nefarious reasons. I thought that was super scary.

I wish I could hear Professor Brailsford's thoughts on AI these days man

What is an irregular expression?

Pong.

3:55 As someone born in the former GDR, I find it cute to label a Trabi as "a car"...

Why is there always so much mess and clutter in the background of these videos? Do you film them in abandoned buildings?

Great video!

Is it stealing to download images off the internet without permisision and use them to ai photobash something together for profit? Yes.

But How does an AI understand comedy? One bit at a time!

Tio Paulo

Although this features lots of catch phrases; it doesn’t include any genuine explanations ( ? ) The thing is, up until very recently; computers were ‘programmed’ by humans, but now - humans program computers to figure things out on their own … & these computers are under no obligation to tell The humans how they’re doing these things, after they become proficient at these things ! Also; these film makers are astounding lazy & they really should have hired some 7_yrolds to do The animation bits !

I haven't tried in a while, but I used to get it to do things it didn't want to, by telling it, I was doing it for a test environment

A very important bit that was skipped over is how you get an LLM to talk about an image (multimodal LLM)! After you got your embedding from the vision encoder you train a simple projection layer that aligns the image embedding with the semantic space of the LLM. You train the projection layer so that the embedding of the vision encoder produces the desired text output describing the image (and or executing the instructions in the image+prompt). You basically project the "thoughts" of the part that sees (the vision encoder) into the part that speaks (the massive LLM).

When they say "high dimensional" in the vector context, I like to imagine it like an RPG character stat sheet, as each independent stat on that sheet can be considered its own dimension.

Me: added to resume ai expert

AI art is great.

Ritchie also wrote some very successful songs in the 80's, often with haunting lyrics: "Hello... was it me you're coding for?"

The biggest problem with unlocking a face with your phone is that you'll laugh too hard to hear the video for a minute or so.

Nice animator

Cat picture needs a tail!

One day, we'll give these models some high resolution images and comprehensive explanations and their minds will be blown! It's astonishing how good even a basic perceptron can be given 28x28 pixel images!

“AI”

5.02 Never seen a "boat wearing a red jumper" before lol

Do any of you ever get this weird feeling that you need to buy Mike a beer? Or perhaps, a substantial yet unknown factor of beers?

13:54 you actually probably can, at least to an extent there’s been some recent research on the idea of going backwards from embeddings to text, maybe look at the paper “Text Embeddings Reveal (Almost) As Much As Text” (Morris et al) the same thing has been done with images from a CNN, see “Inverting Visual Representations with Convolutional Networks” (Dosovitsky et al) neither of these are with CLIP models so maybe future research? (not that it’d produce better images than a diffusion model)

Where did they get those images from? Any copyright infringement?

"Maybe it's security through obscurity" SeemsGood

The exchange at 6:50 made me laugh out loud and I had to show that part of the video to the people around me😆😆

Normally this guy is great for explaining things and super clear, but in this case it feels like he's kind of assuming some prior knowledge or understanding, and not really giving us the 'explain it like I'm five years old' version. And for a subject this complicated, you kind of need that.

'and we want an image of foggonstilz' me: wat 'we want to pass the text of farngunstills' me: u wot m8

You lost me. Far too complicated for this channel.

se han ddo cuenta que estos son un grupito de federicos a los que les patila el cloche.. ellos hacen videos qué no dicen nada

Stable Diffusion needs a CEO BTW ....just saying ... 😅

Oh look, a Mike!

Stable Diffusion 3 = potential topic Optimum workflow strategies using Control Nets, LORAS, VEAs etc....?

Personally I woulda just did the colour comparison by putting the 24bit RGB integer colour into a double (the 64bit fpn type) and divided one by the other. If the result is greater than 0.01 or less than -0.01 then they're not close enough to deem the same overall colour and thus not part of the same facing of a shape. **Edit:** When searching for images it might be better to use simple line path (both a 2d and 3d one) matching the given text of what to search for and compare the shapes identified in the images to those 2 paths. If at least 20% of the line paths matches a shape in the image set then it likely contains that what was searching for. Similarly when generating images the line paths should then traced for producing each image then layered on to one image. Finally for identifying shapes in a given image you just iterate through all stored line paths. I believe this is how our brains conceptualise shapes in the 1st place given how our brains have nowhere to draw shapes to compare to. Instead they just have connections between...cells? neurons? Someone will correct me. Anyways they just have connections between what are effectively physical functions that equate to something like this in C: int neuron( float connections[CHAR_BIT * sizeof(uint)] ); Which tells me the same subshapes share neurons for comparisons which means a bigger shape will likely be just something initial nueron to visit, how many neurons to vist, and what angle to direct the path at to identify the next neuron to visit. In other words every subshape would be able to revisit a previous subshapes neruon/function. There might be an extra value or 2 but I'm no neural expert so a rough guess should be accurate enough to get the ball rolling.

Who is watching now comment

Isn't there an AI that fixed shaky video by now?

SBF, is that you?

These guys are so watchable, and somehow they make an inherently inaccessible subject interesting and easy to follow.