Machine Learning

1 readers

1 users here now

Community Rules:

Be nice. No offensive behavior, insults or attacks: we encourage a diverse community in which members feel safe and have a voice.
Make your post clear and comprehensive: posts that lack insight or effort will be removed. (ex: questions which are easily googled)
Beginner or career related questions go elsewhere. This community is focused in discussion of research and new projects that advance the state-of-the-art.
Limit self-promotion. Comments and posts should be first and foremost about topics of interest to ML observers and practitioners. Limited self-promotion is tolerated, but the sub is not here as merely a source for free advertisement. Such posts will be removed at the discretion of the mods.

founded 1 year ago

MODERATORS

communick@academy.garden

What kind of mathematical foundations are required for conducting research across the vast specialised branches of AI/ML/DL? [D] (alien.top)

submitted 1 year ago by HopeIsGold@alien.top to c/machinelearning@academy.garden

20 comments fedilink hide all child comments

The absolute basic mathematics that is required to understand basic ML/DL are calculus, linear algebra, probability and some convex optimisation. We are all aware of that.

But ML and DL has become a vast field both in breadth and depth. A single person can't understand the field entirely. There are specialistions and sub-specialisations and further more.

If you work in a branch of ML/DL research where some other math fundamentals are needed to understand research papers and do innovative research, can you mention your field of work and the math fundamentals that are required to gain entry into your field?

you are viewing a single comment's thread
view the rest of the comments

[–] patrickkidger@alien.top 1 points 1 year ago (2 children)

I do scientific machine learning, with a particular focus on numerical methods and computational biology.

The other big piece of fundamental mathematics needed is differential equations -- ODEs at least, but ideally also SDEs+PDEs+numerics. (Soapbox moment: I find it kind of unusual how poorly this is taught outside of math/engineering courses, given that it's easily the single most successful modelling paradigm of the past few century.)

Just Know Stuff also has a short list of things I consider worth knowing.

[–] Western-Image7125@alien.top 1 points 1 year ago (2 children)

Ok this is… a lot of stuff. Understand probability through measure theory?

[–] ToxicTop2@alien.top 1 points 1 year ago

Sounds freaking fun to me!

[–] esnfdanwm423rsefte@alien.top 1 points 1 year ago

Try not to get to focused on knowing all of such lists but try to skim at least what seems possible because it's nice to have a toolbox in your head.

Most students that still get hired don't know much of proper code architectures, patterns or code decoupling that is very much essential for proper development but still get to learn on the job. Having been a ML engineer for a couple of years I still haven't picked up a lot of statistics or sometimes even architectures because they have never been relevant to our use cases.

At most companies I have been and interviewed at you are expected to learn, not know. You need to be over a base line for the jobs essentially but you should be substantiate why you are a good learner that can pick up anything. One note to this though is if you only limit your search to the biggest companies with unlimited applicant pools. The baseline will definitely rise for minimum requirements and arbitrary filters will be set up just to get rid of the masses and just interviewing the most notable outliers.

[–] Desmack1@alien.top 1 points 1 year ago (1 children)

What about automatic differentiation too!

[–] KingRandomGuy@alien.top 1 points 1 year ago

This isn't really a fundamental piece of mathematics, it's just an algorithm built on the chain rule.