Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Little Learner: A Straight Line to Deep Learning

mitpress.mit.edu

Post reply on HN
Print view View original

91–100 of 102 posts

Re: The Little Learner: A Straight Line to Deep Learning

#91

Post by barrenko »

hedgehog0 wrote: post #72
havercosine wrote: post #66

Earlier quoted context omitted.

'Little..' series books are famous for their pedagogical style: small bit of concepts, presented as a dialogue, making you think at each step, in a well thought out order. It will appeal to anyone who wants to build it out themselves, understand the 'big' ideas in the field. Skim through the second chapter to see how this will be done: they define line function in inverted way where slope and intercept are taken as p…

So you mean that fast.ai maybe a better resource for people new to general ML?

Probably, but "two Python books" mentioned in this thread are even more suited to a general beginner.

Re: The Little Learner: A Straight Line to Deep Learning

#92

Post by hedgehog0 »

barrenko wrote: post #91
hedgehog0 wrote: post #72

Earlier quoted context omitted.

So you mean that fast.ai maybe a better resource for people new to general ML?

Probably, but "two Python books" mentioned in this thread are even more suited to a general beginner.

Thank you. Maybe it's not modest, but I do not consider myself to be a "general beginner". I majored in math and CS, so I'm thinking about getting something more hard-core and serious...

Re: The Little Learner: A Straight Line to Deep Learning

#93

Post by barrenko »

hedgehog0 wrote: post #92
barrenko wrote: post #91

Earlier quoted context omitted.

Probably, but "two Python books" mentioned in this thread are even more suited to a general beginner.

Thank you. Maybe it's not modest, but I do not consider myself to be a "general beginner". I majored in math and CS, so I'm thinking about getting something more hard-core and serious...

Great. Have you tried just jumping in to Karpathy's video series on youtube? Maybe that plus reading some papers?

Re: The Little Learner: A Straight Line to Deep Learning

#94

Post by legerdemain »

bainsfather wrote: post #83
legerdemain wrote: post #9

Earlier quoted context omitted.

In the same series, there's also "A Little Java" by Matthias Felleisen, one of the creators of Racket (a popular Lisp). https://mitpress.mit.edu/9780262561150/a-little-java-a-few-p...

Is it still good? I see it was published in 1997. I like the style of these books, and might read this one - I'm just worried that Java has changed quite a bit since then. Is it still, say, 90% OK?

It's mostly a novelty today. Felleisen is a programming language theorist, and him writing a little Java tutorial sounds absurd.

The book itself takes you on a journey to reinvent the Visitor pattern, in a version of Java that predates generics. You can probably find a better explainer for the Visitor pattern in Java.

Re: The Little Learner: A Straight Line to Deep Learning

#95

Post by bainsfather »

legerdemain wrote: post #94
bainsfather wrote: post #83

Earlier quoted context omitted.

Is it still good? I see it was published in 1997. I like the style of these books, and might read this one - I'm just worried that Java has changed quite a bit since then. Is it still, say, 90% OK?

It's mostly a novelty today. Felleisen is a programming language theorist, and him writing a little Java tutorial sounds absurd. The book itself takes you on a journey to reinvent the Visitor pattern, in a version of Java that predates generics. You can probably find a better explainer for the Visitor pattern in Java.

Thanks. I see it for sale for £4 second hand, so maybe I should buy it for the novelty value alone. I presume I shouldn't try to refresh my java from it though.

Re: The Little Learner: A Straight Line to Deep Learning

#96

Post by veqq »

bzhang255 wrote: post #26

The initial preview is a little disappointing because it feels like it does a great job displaying the pedagogical style but not enough to demonstrate the pedagogical value (i.e. whether this resource actually does a good job teaching a complex topic). Given the wealth of information and the problem of appraising it all, I don't think it helps this book that it costs $55 and requires the reader to learn an esoteric l…

Checking the library the book implements can help: https://github.com/themetaschemer/malt

Re: The Little Learner: A Straight Line to Deep Learning

#97

Post by leoc »

leoc wrote: post #33
drcode wrote: post #17

Earlier quoted context omitted.

this series of books has an unusual pedagogical style, with a big "bang for the buck" in terms of building up complex systems from basic steps there is zero fluff, almost zero narration the books are basically just input output pairs of "now do this, and now that happens" they are basically a sort of brain data dump for people who can think with computer code

Guy Steele's keynote talk at Dan Friedman's 60th-birthday academic festival, "Dan Friedman--Cool Ideas", gives lots of the background: https://youtu.be/IHP7P_HlcBk?t=2198 . Apparently the format was originally a parody or adaptation of that of an IBM Fortran instruction manual based on Skinnerian programmed-instruction ideas.

You can see the FORTRAN texts on Bitsavers: https://bitsavers.org/pdf/ibm/1130/lang/

Re: The Little Learner: A Straight Line to Deep Learning

#98

Post by havercosine »

barrenko wrote: post #93
hedgehog0 wrote: post #92

Earlier quoted context omitted.

Thank you. Maybe it's not modest, but I do not consider myself to be a "general beginner". I majored in math and CS, so I'm thinking about getting something more hard-core and serious...

Great. Have you tried just jumping in to Karpathy's video series on youtube? Maybe that plus reading some papers?

I second this. Given your background (lisper) maybe do the Little Learner book once it is out + Karpathy's video series. Follow it up by building a slightly complicated application in your favourite domain (text, images, videos, time series).

Also word of advise from my experience (I'm not an expert in DL either): Think of DL field as a game of lego blocks. The ideas in this book / Karpathy's videos are the basic lego blocks: parameterised linear functions, non-linearities, auto-grad, cross-entropy / KL divergence loss and gradient descent. Then there is entire body of more complex legos discovered simply by practice (alchemy!): transformer blocks, layer norm, max-pooling etc. It is impossible to understand how the second kind were obtained from first principles. The trick is to not beat yourself up about the advanced blocks too much but just play around, read up things in papers. Just focus on fundamental blocks.

Re: The Little Learner: A Straight Line to Deep Learning

#99

Post by bilqis »

freilanzer wrote: post #68
fithisux wrote: post #63

Earlier quoted context omitted.

Will do. I have also found that 3 days ago https://github.com/zdimension/hm-infer-scheme fantastic for research

CL is decent in this regard: http://alhassy.com/TypedLisp

I don’t have much experience with CL, but I was always a bit put off by a mixture of different paradigms it has, ie it’s not being a pure or kinda pure functional language.

Re: The Little Learner: A Straight Line to Deep Learning

#100

Post by freilanzer »

bilqis wrote: post #99
freilanzer wrote: post #68

Earlier quoted context omitted.

CL is decent in this regard: http://alhassy.com/TypedLisp

I don’t have much experience with CL, but I was always a bit put off by a mixture of different paradigms it has, ie it’s not being a pure or kinda pure functional language.

That's an advantage for me, tbh.
Post reply on HN