Bagel's hello!

Hi, BagelSticks. Welcome to the forum.

You must share some tips on those social skills of yours with us. You're probably ahead of many of us in that department!

I'm a fan of the old maths, too, though I don't do much by way of actual practice. I like knowing about the various principles and applications. I'm currently reading, "Why Machines Learn: The Elegant Maths Behind Modern AI" by Anil Ananthaswamy. It's a bit dense, but I'm learning lots.

Paleontology is also something I'm interested in. My favourite recent book on the subject was "The Rise and Reign of the Mammals" by Steve Brusatte. It's amazing what a single tooth can tell you about an animal.

Hello! Thank you.
Honestly my social skills just invole long checklists in my head, it can be tedious but it works most of the time! ahaha

I love maths for similar reasons, I like knowing how it works, the principles etc, I find it very satisfying. But doing maths also is something I found that calms me down when needed. I have not read that but it sounds interesting! Would you recommend it?

I have read the paleontolgy book! It is truly an amazing one.

What else do you enjoy? If i may ask

BagelSticks said:

But doing maths also is something I found that calms me down when needed. I have not read that but it sounds interesting! Would you recommend it?

Just finished Why Machines Learn: The Elegant Maths Behind Modern AI by Anil Ananthaswamy. If you are interested in maths and interested in AI, it is very ... interesting, genuinely. You'll learn all about the maths behind various kinds of machine learning, including the deep neural networks trained using back propagation that we all know about these days.

Stephen Hawking wrote about popular science books that, "Each equation halves your sales." In that case, this book either isn't in the "popular" category, or I'm the only person to have bought it! There is probably enough dense information, presented as equations, to allow you to write your own neural networks from scratch! At the same time, it is not an AI text book—the text is admirably descriptive.

My ancient engineering maths knowledge has decayed to near nothing, but I still have some grasp of the principles of things like vectors, derivatives and gradient descent. That grasp seemed to be enough. I could just skip over the equations and read the text and get a good feel for what was going on mathematically inside all the machine learning systems.

There is nothing practical about the book. It mentions that, these days, anyone can implement their own neural networks using Python libraries and a bit of code, but it doesn't do any more than mention it. It sticks to its mission of explaining only the maths behind such things.

I bought the Kindle version when it was on special for 99p, so I feel I got good value for money.

BagelSticks said:

But doing maths also is something I found that calms me down when needed. I have not read that but it sounds interesting! Would you recommend it?

Just finished Why Machines Learn: The Elegant Maths Behind Modern AI by Anil Ananthaswamy. If you are interested in maths and interested in AI, it is very ... interesting, genuinely. You'll learn all about the maths behind various kinds of machine learning, including the deep neural networks trained using back propagation that we all know about these days.

Stephen Hawking wrote about popular science books that, "Each equation halves your sales." In that case, this book either isn't in the "popular" category, or I'm the only person to have bought it! There is probably enough dense information, presented as equations, to allow you to write your own neural networks from scratch! At the same time, it is not an AI text book—the text is admirably descriptive.

My ancient engineering maths knowledge has decayed to near nothing, but I still have some grasp of the principles of things like vectors, derivatives and gradient descent. That grasp seemed to be enough. I could just skip over the equations and read the text and get a good feel for what was going on mathematically inside all the machine learning systems.

There is nothing practical about the book. It mentions that, these days, anyone can implement their own neural networks using Python libraries and a bit of code, but it doesn't do any more than mention it. It sticks to its mission of explaining only the maths behind such things.

I bought the Kindle version when it was on special for 99p, so I feel I got good value for money.