Ask HN: Anyone Considering Getting a Math/Phys Msc/PhD | Hacker News Reader

Ask HN: Anyone Considering Getting a Math/Phys Msc/PhD

12 pointsmarkus_zhang13 comments

Just curious, anyone working in IT wants to get a higher level Math/Phys education just for fun and understanding the universe when you are older (40+) and have some savings?

My plan is to take a few courses up to a rigorous understanding of general relativity.

Multi variable Analysis + Linear Algera(review)-> Advanced Linear Algebra (graduate level)-> Vector Analysis for Tensor;

Also an optional course for introductory Differential Geometry to give me a different approach to general relativity(usually approaching from tensor).

For physics it's Classic Mechanics-> Electromagnetic -> Relativity (special and general)

View on HN

I work as a Deep Learning Research Engineer after graduating college. Worked as a Deep Learning consultant while in college.

My three-year science degree makes me quite ineligible to get into a PhD programme directly. I would definitely like to get a PhD if I could. Chances are slim. I worked for cos all my short career and do not have even one published paper.

Want to migrate to a region with more opportunities first, then after getting my feet on solid economic foundation, I will try hard to get into an actually good PhD programme. But in core Deep Learning.

I was a Physics major in college.

As for Math, I really love it, and started loving it more in more recent times.

And I want to contribute fundamental knowledge to the field of Deep Learning, or, more generally, to AI.

And Math is crucial. I have good fundamentals, but I lack good foundations in topics a Math undergrad would know. So, now I have planned to put down three years of my time- part time for Math study where I will study Real and Complex Analysis, Linear Algebra (Physics requires a good base, already), Topology, Group Theory, Category Theory, Game Theory, advanced calculus, Complex Analysis, Real Analysis, etc. On my own. I have subreddits and Math SE. It is going quite well.

I am following books recommended by Susan Rigetti and Mark Saroufim. And I keep my eyes and ears open for good resources and directions.

I feel I am too old to tolerate college beaurocracies and arbitrary rules. Test, semestars and so on. So much hassle.

My plan was to get a MSc in Math, but I figured that no one (= employers) would care in the end, given my career in software development already. When I realized even I didn't care about the degree, I just stopped.

But I do think the material is interesting. So I want to learn in my spare time. I think that's the best alternative for me, as I don't dream about working in academia etc. I was never a stellar student (in a university setting) anyway.

I want to build a stronger foundation in statistics and probability theory. The goal I have in mind is to try and dabble with systematic trading.

I completed a four-year honours level program at a research university. Math has always been one of my biggest passions, but I wanted some real world work experience before heading into grad school.

Fast-forward four years: I now work in software and, as others have said, I have tried and failed to find anyone who cares that I have studied high-level pure math/theoretical physics. So if I ever want back, it would be purely out of passion, not for any professional gains.

I have considered going back just to fulfill the passion side, but right now that passion does not outweigh the burden of potentially going without an income and paying out of savings to support myself and my family while I get a graduate degree, and I doubt that it ever will. I've done a lot of work to make peace with this fact of life. I have a lot of other passions, and I focus on those instead.

Yeah I can definitely understand that other aspects of life takes priority. Would you consider independent study? Many universities offer such option, which saves money because students don't have to register courses they don't enjoy.

I dropped out of grad school where I focused on math/physics. I was too immature to effectively plan my research and write it up. Also, there's a lot to the institution of academia that I find very distasteful.

The courses were not really important, honestly. As opposed to undergrad, the classes were more provided as resources where you could get expert advice on the course material that applied to your research.

I'm still obsessed with the research, though. Linear algebra in particularly. I have had good success with keeping up with my little research interests by reading relevant new papers and books. I still plan on using what I've learned to advance the world in some way. Being able to work on it without looming deadlines and a shoestring stipend is freeing to say the least.

You really don't need to pursue a degree if you want to keep on learning.

I agree with you. After writing down the title I figured what I need is knowledge, not any degree.

I don't want to get a MSc or PhD, only because I don't think it is worth the time. I like some aspects of academia, but I much prefer learning things by myself.

That said, I have a strong desire to learn math, and I have a copy of Understanding Analysis waiting for me to pick up. I think I'd like to learn Analysis, Linear Algebra, Probability, Graphs... Book suggestions are welcome by the way.

For analysis you can't go wrong with Rudin, especially if you already will have experience from Understanding Analysis. For graph theory I quite liked Diestel, but that may not be the best introductory text.

I'm actually on the same boat. My plan is to take a few courses up to a rigorous understanding of general relativity.

Multi variable Analysis + Linear Algera(review)-> Advanced Linear Algebra (graduate level)-> Vector Analysis for Tensor;

Also a course for introductory Differential Geometry.

For physics it's Classic Mechanics-> Electromagnetic -> Relativity (special and general)

So about 6-8 courses. Some are graduate level but most are of undergraduate level I think.

- Linear Algebra: Gilbert Strang's book + MIT OCW video series. Alternatively/additionally Sheldon Axler's Linear Algebra Done Right. Must watch 3blue1brown series on LinAlg.

- Analysis: Spivak's Calculus (yes). Understanding Analysis is also supposed to be good.

- Probability: Introduction to Probability by Dimitri Bertsekas and John Tsitsiklis.

- Graphs: Plenty nice books. Can also choose GT specific chapters from Discrete Math books.

Edit: If you haven't have formal Math training, you need to learn how to write proofs. For Proofs, I wholeheartedly recommend Jay Cummings's book "Proof: A Long Form Text Book".

I'm considering maybe getting a Georgia Tech online master program. But what you mentioned is also very interesting :)

I am going back to school for chemistry degree

That's interesting. Is there any particular field you would like to pursue?

Reply on news.ycombinator.com