Jump to content

AI: MIT or Berkeley?


Ppp

Recommended Posts

Hi all,

I got in MIT and Berkeley (in AI).

There are a few things to think about; but I am not sure which school would be a better choice for me.

Berkeley: good weather, one of the most famous professor in the field I want to work (senior). Also, probably it is more laid-back and collaboration compared to MIT, because it is at west coast (ppl told me so)

MIT: crappy weather, several well-known professors in the field I want to work (both junior and senior). More competitive environment (east coast). In general, got bigger AI department than Berkeley.

My goal is to stay in academia at this point; and I am not sure how each school is doing for this matter.

Almost everything is equivalent between two schools; but the major difference is coming from: weather (favor to Berkeley), marginal difference in profs' reputation (favor to Berkeley), work environment (not sure but favor to Berkeley), and maybe school reputation in general (favor to MIT)

Let me know your opinions and thank you very much in advance! Also, good luck for all of your applications!!

Link to comment
Share on other sites

I don't see see any difference in reputation of two schools. Both of them are the best. No question about it...

If I were you, I would look into your potential advisors... Every professor has his/her own sub-fields from a micro viewpoint (such as general approaches, applications, relations to other schools/industry etc.) even they are in the same research society. You need to check which would be better fit to you. Also, chemistry is a really important factor.

Link to comment
Share on other sites

Pick Berkeley. I am not sure you'll be laid back at Berkeley. In either place, you will be expected to work very hard! Just look at the influence of Berkeley in CS undergrad curriculum right now. In algorithm, we are using Christo's textbook. In AI, we are using Stuart Russell's textbook. That tells me a lot about the strength of Berkeley in CS. In fact, I don't want to start a war here. I think it is probably even better than Stanford's CS.

Link to comment
Share on other sites

visit both schools and decide for yourself:) I believe both of them have weekend events

Sure. I am wondering under the assumption that I will have similar impressions on both schools (and I have been to both schools several times).

Link to comment
Share on other sites

By the way, if your future goal is professorship in academia, you'd better check what happened to the alumni of the groups (especially, most recent three or four people) and how many awards they've won in top conferences.

Link to comment
Share on other sites

By the way, if your future goal is professorship in academia, you'd better check what happened to the alumni of the groups (especially, most recent three or four people) and how many awards they've won in top conferences.

Does "school" or "field" matter; or does only group matter?

Link to comment
Share on other sites

Ultimately what matters is how many high-quality publications you have if you intend to stay in academia. So you want a supervisor who is pretty active in research. In terms of school, MIT and Berkeley are about equal, but Berkeley has a slight edge.

Link to comment
Share on other sites

I think both are good. But since I've been living in the east coast for a long time and I long for pleasant weather, I prefer Berkeley for a change. Unfortunately, I did not get into Berkeley, so I can only keep dreaming...

For algorithm and theories, no question, Berkeley is the best, especially because Richard Karp, who used to be in University of Washington, is now at Berkeley. Also Christo Papadimitriou is there. Those people discovered many of the NP-hard problems!

For AI, I am not sure about it. AI is pretty broad, depends on what area. If you are in machine learning, Berkeley is good since they have strong theory and statistics people there. For robotics, maybe CMU and MIT are better because they have invested more money in this area. Someone can correct me here...

Link to comment
Share on other sites

How about for Computer Vision or Machine Learning?

Anyone?

For Machine Learning, I would go for CMU. I know Berkeley and Stanford have extremely famous ML guys there. But CMU has a much bigger program and faculties, so it would be a safer choice. What if you go for Berkeley but MJ hires another guy or has no openings? And the job score of last-year CMU PhD graduates in academia was pretty good - one Stanford, one Caltech, and one GATech assistant prof...

For Computer Vision, it would be a really hard decision. Top 4 except Stanford are almost equal, but the prof in Berkeley is slightly more famous in the society, in my opinion...... But post-docs of MIT in previous several years have been extremely good. (Many of them now become professors.)

Link to comment
Share on other sites

For Machine Learning, I would go for CMU. I know Berkeley and Stanford have extremely famous ML guys there. But CMU has a much bigger program and faculties, so it would be a safer choice. What if you go for Berkeley but MJ hires another guy or has no openings? And the job score of last-year CMU PhD graduates in academia was pretty good - one Stanford, one Caltech, and one GATech assistant prof...

For Computer Vision, it would be a really hard decision. Top 4 except Stanford are almost equal, but the prof in Berkeley is slightly more famous in the society, in my opinion...... But post-docs of MIT in previous several years have been extremely good. (Many of them now become professors.)

I thought Kanade is the most famous in CV. Which prof from Berkeley did you mean?

Link to comment
Share on other sites

For Machine Learning, I would go for CMU. I know Berkeley and Stanford have extremely famous ML guys there. But CMU has a much bigger program and faculties, so it would be a safer choice. What if you go for Berkeley but MJ hires another guy or has no openings? And the job score of last-year CMU PhD graduates in academia was pretty good - one Stanford, one Caltech, and one GATech assistant prof...

For Computer Vision, it would be a really hard decision. Top 4 except Stanford are almost equal, but the prof in Berkeley is slightly more famous in the society, in my opinion...... But post-docs of MIT in previous several years have been extremely good. (Many of them now become professors.)

Is there any ML profs in MIT, any well-known ones? I thought Berkeley might be a good choice because of MJ and Jitendra Malik; but how about MIT? I know there are Antonio Torralba and Bill Freeman, but how about ML? I am also trying to decide between Berkeley, CMU, and MIT.

Link to comment
Share on other sites

A list of big guys that I read about:

Database: Ullman (Stanford)

Machine Learning: Mitchell (CMU), Moore (CMU), Jordan(Berkeley)

Bioinformatics: Pevner (UCSD), Haussler (UCSC), Tidor(MIT), Berger(MIT), Gerstein(Yale)

Operating Systems: Silberchatz(Yale)

Algorithm: Knuth (Stanford), Karp (Berkeley), Papadimitriou (Berkeley)

Computational linguistics: Fillmore(Berkeley), Hirst(Toronto)

(My area is bioinformatics, that's why its list is very long).

Link to comment
Share on other sites

I guess Tommi Jaakkola is one of famous profs at MIT. But I think ML of MIT is not that strong compared to the other Top4....

The professors bernard listed out is quite accurate, but it may be a little bit outdated. Some of them are now seniors and have some duties in the organizations, so maybe not that active in research now...

Link to comment
Share on other sites

True, leogk. Some of the professors I listed are not active in research: some shift attention to writing books now (Knuth, for example). Some are emeritus professors. But regardless whether they take students or not, their names tell us what area the school is good in. Likely there will be younger profs who are working in the same area.

Seriously, a large percentage of graduates from Berkeley CS later got professorship at top schools. The school name is already enough a reason to go there.

Link to comment
Share on other sites

True, leogk. Some of the professors I listed are not active in research: some shift attention to writing books now (Knuth, for example). Some are emeritus professors. But regardless whether they take students or not, their names tell us what area the school is good in. Likely there will be younger profs who are working in the same area.

Seriously, a large percentage of graduates from Berkeley CS later got professorship at top schools. The school name is already enough a reason to go there.

How about MIT? Don't MIT graduates stay in academia (as many as Berkeley)? I am trying to persuade myself for MIT, but many people (on this thread) say Berkeley is a better choice. Am I right?

Thanks for answering my question! :)

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

This website uses cookies to ensure you get the best experience on our website. See our Privacy Policy and Terms of Use