Shoutout to my college professor who just made his machine learning book free!
I particularly like the book in that it explains the pure computer science and mathematical ideas behind machine learning, model building and validation, without relying on any particular software package. If you are looking for a book that conveys knowledge rather than a glorified tutorial on machine learning, this is it.
Each chapter is written with incremental difficulty, so that one can read up until (hopefully a bit beyond) his/her maths comfort zone.
The book is also entertaining, if anything just read the first paragraph of each chapter and look at the first illustrations!
I am sorry, but I really do not like this style of naming and self-advertising. Is this common in ML?
"LION is a complex array of mechanisms, like the engine in a car, but the user (driver) does not need to know the inner workings of the engine in order to realize its tremendous benefits. LION’s adoption will create a prairie fire of innovation which will reach most businesses in the next decades."
1. Why is it called LION? There is no reference to a lion anywhere in the introduction. Just because the (real) title contains the letters to form lion is (in my opinion) not good enough (machine Learning plus Intelligent OptimizatioN). I hoped we were done with this style 5 years ago.
2. Even before trying to attempt to explain exactely what LION is, we have this car engine analogy. Why not explaining what it is so that I do understang what you are talking about. This just feels like an attempt to obscure complexity. And in (some bad papers) this is a sign that the authors did not fully grasp the matter and can hence not summarize it.
3. Why this bold claim that this will change innovation and business? If the book holds what it promises it will do that on its own.
To be fair, I haven't read the book but will probably do it. And it might be awesome. But these three things raise red flags for me. I read a lot of papers during my PhD and these are three indicators for bad papers. And the authors of those papers were PhD students will oversell there stuff anyways. But this is a book. Please do not do that...
How is "free" relevant here? The underlying truth is that price in general is not necessarily correlated with quality. That would be true if the book were free, or if it cost $20.00, $200.00 or $2,000.00.
IOW:
If he is selling this for $200.00, how worthwhile is it?
Sorry to be skeptical, but expensive doesn't always equal great quality.
Typically, value is entirely the realm of the receiver. Many things are free and incalculably valuable. Start reading the book and see if you value it.
Did not downvote. You asked a valid question that furthered the discussion
Do you have the same concerns and vocalize them when you get gifts on christmas, also?
He payed the price to write this book (and maybe got salary for his work), so you can have it for free.
Sorry, but your comment seems awfully misguided.
(assuming you celebrate christmas)
EDIT: instead of dismissing it directly, how does actually reading some parts of it first sound like?
Ah, and dismissing an important argument with "excluding open source" is not a good way to actually find truth. Because if excluding this argument means changing the question and scrambling the context too much, the answer to this modified questions does not reflect reality that much: https://www.youtube.com/watch?v=Qc7HmhrgTuQ
I particularly like the book in that it explains the pure computer science and mathematical ideas behind machine learning, model building and validation, without relying on any particular software package. If you are looking for a book that conveys knowledge rather than a glorified tutorial on machine learning, this is it.
Each chapter is written with incremental difficulty, so that one can read up until (hopefully a bit beyond) his/her maths comfort zone.
The book is also entertaining, if anything just read the first paragraph of each chapter and look at the first illustrations!
Additional info here: http://intelligent-optimization.org/LIONbook