/sci/ - Science & Math

File: 436 KB, 609x532, 1575524542777.png [View same] [iqdb] [saucenao] [google]

Anonymous Mon Dec 30 11:29:34 2019 No.11264421 [Reply] [Original]

Hey lads I am a pure math PhD with good programming and computer science experience. I have lots of free time during the holidays now and I want to learn more about machine learning and AI, by doing practical projects. I have already done the basic character recognition stuff and learned the theory behind neural nets, PCA and other popular topics. What could I work on next?

>>	Anonymous Mon Dec 30 11:44:13 2019 No.11264443 >>11264421 > math PhD > 300 mg > any medication you want

Anonymous Mon Dec 30 12:51:18 2019 No.11264565

>>11264421
GANs and variational autoencoders are extremely interesting in their properties. t-sne is a great visualization tool, albeit a bit unforeseeable.
Then also Berry numbers as applied in geometrical data analysis. They give you topological information about your data which can hint at unexpected connections.

>>	Anonymous Mon Dec 30 13:12:48 2019 No.11264618 >>11264443 >sensible_chuckle.jpeg2

>>	Anonymous Mon Dec 30 13:28:24 2019 No.11264644 latent dirichlet allocation and natural language processing in general seems like a good next step

>>	Anonymous Mon Dec 30 13:32:37 2019 No.11264649 >math >phd So tell us anon you had more than enough Time to figure out the best way to LEARN MATH. What is it?

Anonymous Mon Dec 30 13:32:55 2019 No.11264650

>>11264421
http://www.deeplearningbook.org/

Statistical learning theory is based on functional analysis, SLT was base theory for SVM, but translate to modern Deep Learning technics,

The Nature of Statistical Learning Theory Vladimir Vapnik and other book(Statistical Learning Theory) from same autor, a lot SLT is used in deep learning

Algebraic topology for Machine Learning, UMAP as advance t-SNE as non-linear PCA is based on AT
https://medium.com/@dan.allison/dimensionality-reduction-with-umap-b081837354dd..
Book 300 page from INRIA
https://geometrica.saclay.inria.fr/team/Fred.Chazal/papers/CGLcourseNotes/main.pdf

Math on ML
https://mathematical-coffees.github.io/listing/
https://mathematical-coffees.github.io/slides/mc11-oyallon.pdf

Optimal transport theory
https://mathematical-coffees.github.io/slides/mc01-courty.pdf

>>	Anonymous Mon Dec 30 13:38:58 2019 No.11264659 >>11264421 How's your game theory? Work on this if you got game. https://youtu.be/cD-eXjf854Q

>>	Anonymous Mon Dec 30 13:49:27 2019 No.11264679 >>11264650 https://mathematical-tours.github.io/book/ http://www.numerical-tours.com/

>>	Anonymous Mon Dec 30 18:12:12 2019 No.11265172 File: 1.37 MB, 1140x4777, official mg curriculum.png [View same] [iqdb] [saucenao] [google] >>11264649 >So tell us anon you had more than enough Time to figure out the best way to LEARN MATH. What is it?

>>	Anonymous Mon Dec 30 19:01:27 2019 No.11265287 >>11264650 Nice list. Thank you, Anon.

Advanced search
Text to find
Subject [?]Search by post subject. Leave empty for any.
Username [?]Search for user name. Leave empty for any user name.
Tripcode [?]Search for tripcode. Leave empty for any.
Email [?]Search by email. Leave empty for any.
Filename [?]Search by image filename. Leave empty for any.
From Date [?]Enter what date to start searching from. Format is YYYY-MM-DD
To Date [?]Enter what date to start searching until. Format is YYYY-MM-DD
Image hash
Search in	All Posts OPs Only
Deleted posts	Show all posts Show only deleted posts Only show non-deleted posts
Internal posts	Show all posts Show only internal posts Show only archived posts
Order	New posts first Old posts first
Capcode	All Posts Only by Users Only by Mods Only by Admins Only by Developers
Results	Posts Threads
Action	[ Simple ]