Beautiful mind

What is a reasonable classic respectable algo to use this situation ? new stuff The point is to generalize beyond our data :)))😍

find patterns that are not there :))

Summary of all textbooks in one video to understand ML . Thank you , Cassie

I assume you have some reasonable math background, but your assertion that “a Bayesian can never be wrong, because it is just their own opinion” completely misses the point and application of it, because it transcends beliefs in how we use it from everything to bioinformatics to quantum mechanics. For example, a Bayesian can leverage a Markov Chain Monte Carlo - which is in effect randomly sampling and walking around the space of possibilities. It’s analogous to the frequentists approach of using permutation to get an effective null. What really matters are the factors affecting accuracy: 1. Prior information: When strong, reliable prior information exists, Bayesian methods can outperform frequentist approaches by incorporating this knowledge. 2. Model complexity: For complex models with many parameters, Bayesian methods often have an advantage due to their flexibility in handling uncertainty. 3. Nature of the problem: Some questions are more naturally framed in Bayesian terms (e.g., "What's the probability this hypothesis is true?"), while others fit better with frequentist thinking and simplifying assumptions. 4. Sample size: With large samples, both approaches often converge to similar results. Frequentist methods may have a performance edge here due to their focus on long-run performance. 5. Assumptions: Both approaches rely on assumptions. The approach that better matches the true underlying process will tend to perform better.

I don't think there's really an identity of being one or the other. This framing seems to me to lend permission to making perspective equally authoritative. It's a laundering of authority into being that isn't there

A danger exists in drawing from the same poisoned well made up of people cognitively bad at probability.

The Baeysian recognizes that the probability is relative to perspective which means recognizing the context of other's perspevtive. Perspective must involve uncertainty. The problem is that it usually loses that component. I think some people combine the two to eliminate their feeling of discomfort.

Most people: picks one of them Me: gets existential about both options and my original 50% being i am watching a pre-recorded video and what chance means and how it is when we have expectations on chance😅 Idk what category i fit into now

I sounds like you're saying that Bayesians just pull opinions out of their butts. I work in AI and it is well known (mathematically proven) that the Bayesian update is the optimal way to add new data to your existing knowledge. In our world, we have very few certainties, as a result, we are forced to make guesses, estimate things, and incorporate uncertain data into an uncertain pool of knowledge. The optimal way to learn is demonstrably Bayesian, so in the field of AI, there is no alternative.

You’re beautiful

Does anyone know where video 001 to 056 went? Did Cassie set them to private?

Amazing.. The summary of the all books about ML :))))

Your video is private . No allow to watch :(

Any clue what happened to the other videos from the playlist?

have you guys watch AI learn how to walk? it mostly brute force and takes a lot of time and the result are mostly hilarious. yeah I don't think AI will take over anytime soon, but maybe far in the future

Cassie my prior about you being a frequentist is 90%

I still wonder about Granite models, given their own Merlinite trained versions are much more performant.

A vlog by one of my favourite teachers! Loved your MFML series..

"a toilet"

Best Notification, 🥰

Wow sounding powerful the Great Cassie !!

Thanks Cassie K desde Colombia

Noone is ever clear about what it is we will actually do. For work, I mean not a hobby. Much research shows that we thrive when we strive. We need purpose. Challenges. Again, she says words but nothing concrete...

1:17 - at last, someone dared to combine math with ballet dancing!

Damnit you anticpiated my smartass comment that coins aren't perfectly balanced XD

Thank you very much! I suppose I've finally got it.

You said at the beginning you would flip the coin until it turned up heads, and that sounded like it was going to be interesting. And then you dwelled on exactly that one coin flip. It would have been more interesting if you spoke more instead about how many more coin flips might be necessary from a data set before you get 50 heads. This would be evaluated from the already-gathered coin flip data. I feel like a single data point of a single coin flip isn't a useful frame when you're anywhere close to statistics.

my god!!!

I'm looking for the full 10-hour course but can't find it anywhere. Does it still exist?

The irony will be ai seeing humans as an inneficiency. Then what?

a beautiful vlog in a beautiful city presented by a beautiful host.. entertaining as always :)

Thank you Cassie. I finally deeply understood H0

the subtle humor in between makes the course so much more engaging.. One of the greatest instructors i've ever seen..i can't believe i completed a 1.5 hr video in one go.. kudos to Cassie, admirable job..

It's a domain which really needs a lot of research, good work!

Is midjourney integrated into photoshop? Or is that a photoshop feature?

Who knows the coin might still be flipping inside her palm untill i see the result.

All true but it won’t occur as long as AI is owned by a mega corp.

Thanks for this.

I love youuuuuuuuuu

Very insightful talk, improve my vision about the topic, thanks

Many ways that this is wrong. Sorry but Bayesian stat does not give a whit about perspective. Only priors and posteriors.

Jem Corcoran, A math professor at Colorado University says probability is about the future and statistics is about the past. I think, frequentists will say that one of the pasts either happened or didn't happen and there is no probability and they have nothing to say about the past. Am I correct? The random variables, is it a frequentist idea or a Bayesian idea or some-other-statistian's-name-ian idea?

Is it correct to say all frequentists will have the same answer given the same data and all Bayesians will have the same answer given the same data and the same initial beliefs?

Bayesians may have a notion of error. In fact they should at least based only on your video. If I ask a Bayesian what the error is, they should have an opinion about it. To say there is no error is frequentist because they feel the error is 0 or 1. Since they can't know for sure, there don't want to concern themselves about it. I am not sure I am right about the third last sentence of the paragraph. Maybe, if we ask about their opinion about the probability that error is greater than some percentage, they should have an answer, I believe. But, there are 2 types of Bayesians. One kind that uses Bayesian methods to calculate probabilities. Other kind that actually believes that the mean of a population or some other statistics is actually random.

Hi & thank you for that motivational tutorial. Question: how do I correctly specify identify the distribution of the data to simulate for univariate, bivariate and multivariate data situations?

Thank you for your clear explanation!

I like to use the metaphor "AI is a unicorn in a China shop."

Thank you for saving my brain cells that were about to be destroyed by watching several other videos on p-values 〒▽〒

holy sh this is the best video i've watched im taking ap psych and i had no clue what this meant. thank yo uso much