1/ I’ve thought for many years that strong verbal skills probably contribute to strong coding skills. “Code should read like a book!” was something a mentor told me that really stuck. Perhaps surprisingly- I’m not great with math- especially linear algebra of all things.https://twitter.com/fchollet/status/1208079432373563392 …
-
Show this thread
-
Replying to @citnaj
Not great with maths? This statement blows my mind a bit. I mean, Im guessing you understand the below shenanigans?pic.twitter.com/U2hEbAfBoT
2 replies 0 retweets 1 like -
Replying to @pastaduck
I’m motivated to point all this out because I think the importance of math prerequisites is a bit overemphasized and might be keeping a lot of otherwise talented people from giving the fields a try.
2 replies 0 retweets 1 like -
Replying to @citnaj
Linear algebra and calculus an important pre requisite in your opinion? Or do you reckon you can get pretty far just by applying and building upon code in the wild.
2 replies 0 retweets 0 likes -
Replying to @pastaduck
1/ I don’t think either are required and agree with
@fastdotai that algebra suffices. The core concepts from linear algebra and calculus that are most useful to know are very easy to get the basic gist on quickly (derivatives and matrix multiplication).1 reply 0 retweets 1 like -
2/ My personal approach to figuring stuff out really boils down to a combination of leaning heavily on experimentation, holding all ideas of how things as temporary and subject to change, sprinkle in cross pollination from other domains/papers, and good old fashioned problem
1 reply 0 retweets 2 likes -
Hmm, I'd say a solid understanding of linear algebra and calculus are required; it saves time in the long run to be able to identify what math is decoration and what is essential. Having access to lower level building blocks also allows for faster understanding, recall.
2 replies 0 retweets 2 likes -
Replying to @sir_deenicus @citnaj and
The risk of taking a code heavy approach, and I see this a lot, is you end up thinking in terms of frameworks instead of concepts. For example, rather than decomposing tasks in terms of Pytorch functions, or worse/higher--BERT--you ask, what is self-attention really doing?
2 replies 0 retweets 4 likes -
Replying to @sir_deenicus @citnaj and
How is a Transformer simulating a graph? How can I build a language model with random vectors yet respect word order? Being able to lean on say, linear algebra really helps a lot. Problem is math is taught as if you have to be smart instead of something which makes you smarter.
2 replies 0 retweets 3 likes -
Replying to @sir_deenicus @citnaj and
I know there are a lot of people who like math because they like puzzles and are clever and math is optimized for them. But I hate puzzles; I like math because I get to lean on a thousand year old knowledge base of solved problems. It's sort of a type system for thinking.
1 reply 0 retweets 2 likes
Now see on that last idea of it being a type system for thinking.... I’ve called mathematical notation “a really shitty language” in the sense that there’s a whole bunch of overloads of the same one (one!) character symbols and it fails at expressing algorithms effectively.
-
-
I completely agree (I also think probability theory is the worst offender by far; in papers, this crops up with Expectation notation a lot, often not even consistent throughout same paper). It's understandable given its age but hostile to learners & should def be improved.
1 reply 0 retweets 0 likes -
Replying to @sir_deenicus @citnaj and
An example of what I mean by augmenting is: I've sometimes found myself needing to work out possible combinations. Most who don't know math struggle to do this while all I had to do was work out what type of permutation calculation applied. It's not at all that I was smarter!
0 replies 0 retweets 2 likes
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.