A book on machine learning

Discussion of chess software programming and technical issues.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Look
Posts: 238
Joined: Thu Jun 05, 2014 12:14 pm
Location: Iran
Full name: Mehdi Amini
Contact:

A book on machine learning

Post by Look » Sun Oct 06, 2019 6:28 pm

Hi,

I studied the following book. It's codes are written in JavaScript , Python , C++ . The book is a recent and fine introduction on several AI methods.

Here is the book link.

Rémi Coulom
Posts: 434
Joined: Mon Apr 24, 2006 6:06 pm
Contact:

Re: A book on machine learning

Post by Rémi Coulom » Mon Oct 07, 2019 12:15 pm

I took a quick look at the content, and it seems there is extremely little machine learning in this book, except for a decision tree in Chapter 2. Mostly genetic algorithms and population-based methods. No neural network. But what they describe may be fun programming experiments.

brianr
Posts: 356
Joined: Thu Mar 09, 2006 2:01 pm

Re: A book on machine learning

Post by brianr » Mon Oct 07, 2019 1:11 pm

I think this one is pretty good, albeit for Go not Chess, and it's a bit pricey.
However, it does cover quite a lot specifically for Alpha Zero, and the code code samples do work (with a bit of tweaking last time I tried).

https://www.amazon.com/Deep-Learning-Ga ... 1617295329

User avatar
asanjuan
Posts: 211
Joined: Thu Sep 01, 2011 3:38 pm
Location: Seville, Spain

Re: A book on machine learning

Post by asanjuan » Mon Oct 07, 2019 1:24 pm

This is very good.
"Make Your Own Neural Network" by Tariq Rashid.
A good starting point. It doesn't cover Convolutional NN but all the basics are there.

Daniel Shawul
Posts: 3758
Joined: Tue Mar 14, 2006 10:34 am
Location: Ethiopia
Contact:

Re: A book on machine learning

Post by Daniel Shawul » Mon Oct 07, 2019 2:05 pm

where my free ebooks at?
There is one free deep learning book i know of
https://github.com/janishar/mit-deep-le ... ngbook.pdf

fabianVDW
Posts: 73
Joined: Fri Mar 15, 2019 7:46 pm
Location: Germany
Full name: Fabian von der Warth

Re: A book on machine learning

Post by fabianVDW » Mon Oct 07, 2019 4:30 pm

Rémi Coulom wrote:
Mon Oct 07, 2019 12:15 pm
I took a quick look at the content, and it seems there is extremely little machine learning in this book, except for a decision tree in Chapter 2. Mostly genetic algorithms and population-based methods. No neural network. But what they describe may be fun programming experiments.
AFAIK genetic algorithms definitly belong to the term "machine learning". Aside from that, genetic algorithms combined with neural networks (mostly NEAT and variations) are an interesting research point, albeit I haven't seen the usage for chess.
Author of FabChess: https://github.com/fabianvdW/FabChess
A UCI compliant chess engine written in Rust.
FabChessWiki: https://github.com/fabianvdW/FabChess/wiki
fabianvonderwarth@gmail.com

Richard Allbert
Posts: 767
Joined: Wed Jul 19, 2006 7:58 am

Re: A book on machine learning

Post by Richard Allbert » Thu Oct 10, 2019 6:27 am

asanjuan wrote:
Mon Oct 07, 2019 1:24 pm
This is very good.
"Make Your Own Neural Network" by Tariq Rashid.
A good starting point. It doesn't cover Convolutional NN but all the basics are there.
I'll second this, for a non academic scatter brain like me, this book explained things really well. I've managed to complete a few home projects since.

Another book that follows up from the above one is A Simple Guide on Machine Learning with Neural Networks - this is short, sweet, and adds bias nodes and things like that.

https://www.amazon.de/Machine-Learning- ... 679&sr=8-6

I find the concept fascinating - whether it's making a machine learn to play space invaders, or an autonomous raspberry pi car, it's amazing to watch the 'learning' process.

User avatar
towforce
Posts: 9298
Joined: Wed Mar 08, 2006 11:57 pm
Location: Birmingham UK

Re: A book on machine learning

Post by towforce » Sat Oct 19, 2019 3:09 pm

You can have a play with a small NN for your own training here - link.

That page suggests that if you want to take things further, have a go with the popular TensorFlow library, which has no been ported into many languages.
Love of truth is the best defence against ideological possession.

Post Reply