John Urschel: Buffalonian, Baltimore Raven, MIT Math Professor

[hondo in seattle]

Urschel grew up in Buffalo, graduated from Canisius, became a Baltimore Raven, and just got a job as a math professor at MIT.  He wrote his Ph.D. thesis on "Graphs, Principal Minors, and Eigenvalue Problems" and has published over a dozen peer-reviewed papers.  All by the age of 32.  

 

Makes me feel like I haven't tried hard enough.  

[Punching Bag]

6 minutes ago, hondo in seattle said:

Urschel grew up in Buffalo, graduated from Canisius, became a Baltimore Raven, and just got a job as a math professor at MIT.  He wrote his Ph.D. thesis on "Graphs, Principal Minors, and Eigenvalue Problems" and has published over a dozen peer-reviewed papers.  All by the age of 32.  

 

Makes me feel like I haven't tried hard enough.  

 

I was an honors math major in college and read briefly the paper.

https://math.mit.edu/~urschel/publications/thesis.pdf

 

I agree with you I did not work hard enough but I got tired of school taking whatever work I could to pay bills.  At one point I was being chased out of lounges for sleeping there and sleeping in office at my job since I had no housing.

https://dspace.mit.edu/handle/1721.1/140006

[The Wiz]

I can't even pronounce the title of his thesis.

[LabattBlue]

5 minutes ago, The Wiz said:

I can't even pronounce the title of his thesis.

Let alone understand what it means. 😂

Edited October 6, 2023 by LabattBlue

[Big Turk]

40 minutes ago, hondo in seattle said:

Urschel grew up in Buffalo, graduated from Canisius, became a Baltimore Raven, and just got a job as a math professor at MIT.  He wrote his Ph.D. thesis on "Graphs, Principal Minors, and Eigenvalue Problems" and has published over a dozen peer-reviewed papers.  All by the age of 32.  

 

Makes me feel like I haven't tried hard enough.  

 

Very rare for a person to be gifted with elite traits that allow them to reach the pinnacle of their profession in multiple divergent areas...

[hondo in seattle]

5 minutes ago, LabattBlue said:

Let alone understand what it means. 😂

 

If you don't want to read the whole thesis, Urschel provided a short summary:

 

"This thesis considers four independent topics within linear algebra: determinantal point processes, extremal problems in spectral graph theory, force-directed layouts, and eigenvalue algorithms. For determinantal point processes (DPPs), we consider the classes of symmetric and signed DPPs, respectively, and in both cases connect the problem of learning the parameters of a DPP to a related matrix recovery problem. Next, we consider two conjectures in spectral graph theory regarding the spread of a graph, and resolve both. For force-directed layouts of graphs, we connect the layout of the boundary of a Tutte spring embedding to trace theorems from the theory of elliptic PDEs, and we provide a rigorous theoretical analysis of the popular Kamada-Kawai objective, proving hardness of approximation and structural results regarding optimal layouts, and providing a polynomial time randomized approximation scheme for low diameter graphs. Finally, we consider the Lanczos method for computing extremal eigenvalues of a symmetric matrix and produce new error estimates for this algorithm."

 

Typical locker room jock talk.  

[TheBrownBear]

Great story!  Congrats to John!

 

Also, it's pretty funny to think about going from a profession where your colleagues are guys like Tyreek Hill to becoming an MIT professor. 

[The Jokeman]

1 hour ago, hondo in seattle said:

Urschel grew up in Buffalo, graduated from Canisius, became a Baltimore Raven, and just got a job as a math professor at MIT.  He wrote his Ph.D. thesis on "Graphs, Principal Minors, and Eigenvalue Problems" and has published over a dozen peer-reviewed papers.  All by the age of 32.  

 

Makes me feel like I haven't tried hard enough.  

and I never heard of him before today. 

[Mr. WEO]

Original title:

 

"Graphs, Principal Minors, and Eigenvalue Problems and Lamar's Wonderlic Score"

 

 

sent back for revision...

[WhoTom]

1 hour ago, hondo in seattle said:

Makes me feel like I haven't tried hard enough.  

 

I'm proud of my CV as it stands, but I also recognize that MIT professors and their ilk are in a totally different league, so I try not to compare myself to them.

 

 

[boyst]

1 hour ago, hondo in seattle said:

Urschel grew up in Buffalo, graduated from Canisius, became a Baltimore Raven, and just got a job as a math professor at MIT.  He wrote his Ph.D. thesis on "Graphs, Principal Minors, and Eigenvalue Problems" and has published over a dozen peer-reviewed papers.  All by the age of 32.  

 

Makes me feel like I haven't tried hard enough.  

by that same age Travis Henry had 14 children.

 

this guy hasn't even done that much.

[LabattBlue]

1 hour ago, hondo in seattle said:

 

If you don't want to read the whole thesis, Urschel provided a short summary:

 

"This thesis considers four independent topics within linear algebra: determinantal point processes, extremal problems in spectral graph theory, force-directed layouts, and eigenvalue algorithms. For determinantal point processes (DPPs), we consider the classes of symmetric and signed DPPs, respectively, and in both cases connect the problem of learning the parameters of a DPP to a related matrix recovery problem. Next, we consider two conjectures in spectral graph theory regarding the spread of a graph, and resolve both. For force-directed layouts of graphs, we connect the layout of the boundary of a Tutte spring embedding to trace theorems from the theory of elliptic PDEs, and we provide a rigorous theoretical analysis of the popular Kamada-Kawai objective, proving hardness of approximation and structural results regarding optimal layouts, and providing a polynomial time randomized approximation scheme for low diameter graphs. Finally, we consider the Lanczos method for computing extremal eigenvalues of a symmetric matrix and produce new error estimates for this algorithm."

 

Typical locker room jock talk.  

To quote the scarecrow…If I only had a brain.  

[Don Otreply]

And?

 

 

 

 

 

 

 

 

 

 

 

 

😂

[US Egg]

2 hours ago, hondo in seattle said:

"This thesis considers four independent topics within linear algebra: determinantal point processes, extremal problems in spectral graph theory, force-directed layouts, and eigenvalue algorithms. For determinantal point processes (DPPs), we consider the classes of symmetric and signed DPPs, respectively, and in both cases connect the problem of learning the parameters of a DPP to a related matrix recovery problem. Next, we consider two conjectures in spectral graph theory regarding the spread of a graph, and resolve both. For force-directed layouts of graphs, we connect the layout of the boundary of a Tutte spring embedding to trace theorems from the theory of elliptic PDEs, and we provide a rigorous theoretical analysis of the popular Kamada-Kawai objective, proving hardness of approximation and structural results regarding optimal layouts, and providing a polynomial time randomized approximation scheme for low diameter graphs. Finally, we consider the Lanczos method for computing extremal eigenvalues of a symmetric matrix and produce new error estimates for this algorithm."

That’s what I thought it was about.

[Punching Bag]

2 hours ago, Big Turk said:

 

Very rare for a person to be gifted with elite traits that allow them to reach the pinnacle of their profession in multiple divergent areas...

 

There actually have been a number of doctors and other people with elite traits who decided to go to NFL first but did not major in basket weaving in college.

[WhoTom]

42 minutes ago, Don Otreply said:

And?

 

😂

 

Several years ago, one of my colleagues interviewed for a professor's position at Purdue University. (He and I taught at a community college, and he had a side gig as an engineer because he was a workaholic.) The interviewer asked him about publications and he mentioned the textbook he'd written.

 

Unfazed, the interviewer responded, "So that's just the one textbook?"

 

Then he asked about any other honors of note and my friend said, "I have a patent in my name."

 

Predictably, the interviewer followed up with, "So that's just the one patent?"

 

Sheesh.

 

🙂

 

 

 

[WhoTom]

3 hours ago, The Wiz said:

I can't even pronounce the title of his thesis.

 

When I was working on my dissertation, my son came into my office and noticed the title of one of the papers I was reading. He said, "I see you're studying a foreign language."

 

Sometimes it felt that way.

[Punching Bag]

2 hours ago, WhoTom said:

 

I'm proud of my CV as it stands, but I also recognize that MIT professors and their ilk are in a totally different league, so I try not to compare myself to them.

 

My first submission for publication was when I was 6 years old over 50 years old.  I was reading Scientific American and each issue had one (or more) unsolved problem.  I got a letter back from them "Mr. Host thank you for your submission. We unfortunate publish only first solution.  Dr. xYx and Dr. xYx (do not remember names) submitted solution first. Thank you for your submission and it was actually shorter and more concise than submission published."   They did not even recognize they were sending a letter to a 6 year old kid!

 

I have a paper DISA uses for last 20 years as their standard document for testing and installing guardians (guardians are firewalls used as gateways between connected systems of different security level).

 

I have a patent on OCR licensed by 13 companies which is basis of scanning documents and converting to text.

Apparatus and method of fusing the outputs of multiple intelligent character recognition (ICR) systems to reduce error rate (using fuzzy logic)

I was brought in as system administrator to process hard written documents for government agency to determine best algorithm producing lowest error rate,  While doing this I noticed patterns on failures and I was able to find a way to combine algorithms to radically reduce error rate on tests determining correct letter which neither of original test algorithms could.

 

My issue is not recognition of deeds but the in-between times where I do a lot of drudgery work.

 

[djp14150]

37 minutes ago, Limeaid said:

 

My first submission for publication was when I was 6 years old over 50 years old.  I was reading Scientific American and each issue had one (or more) unsolved problem.  I got a letter back from them "Mr. Host thank you for your submission. We unfortunate publish only first solution.  Dr. xYx and Dr. xYx (do not remember names) submitted solution first. Thank you for your submission and it was actually shorter and more concise than submission published."   They did not even recognize they were sending a letter to a 6 year old kid!

 

I have a paper DISA uses for last 20 years as their standard document for testing and installing guardians (guardians are firewalls used as gateways between connected systems of different security level).

 

I have a patent on OCR licensed by 13 companies which is basis of scanning documents and converting to text.

Apparatus and method of fusing the outputs of multiple intelligent character recognition (ICR) systems to reduce error rate (using fuzzy logic)

I was brought in as system administrator to process hard written documents for government agency to determine best algorithm producing lowest error rate,  While doing this I noticed patterns on failures and I was able to find a way to combine algorithms to radically reduce error rate on tests determining correct letter which neither of original test algorithms could.

 

My issue is not recognition of deeds but the in-between times where I do a lot of drudgery work.

 

What Klein bottle did you fall into on your random walk around the block?
 

[Buffalo716]

4 hours ago, hondo in seattle said:

 

If you don't want to read the whole thesis, Urschel provided a short summary:

 

"This thesis considers four independent topics within linear algebra: determinantal point processes, extremal problems in spectral graph theory, force-directed layouts, and eigenvalue algorithms. For determinantal point processes (DPPs), we consider the classes of symmetric and signed DPPs, respectively, and in both cases connect the problem of learning the parameters of a DPP to a related matrix recovery problem. Next, we consider two conjectures in spectral graph theory regarding the spread of a graph, and resolve both. For force-directed layouts of graphs, we connect the layout of the boundary of a Tutte spring embedding to trace theorems from the theory of elliptic PDEs, and we provide a rigorous theoretical analysis of the popular Kamada-Kawai objective, proving hardness of approximation and structural results regarding optimal layouts, and providing a polynomial time randomized approximation scheme for low diameter graphs. Finally, we consider the Lanczos method for computing extremal eigenvalues of a symmetric matrix and produce new error estimates for this algorithm."

 

Typical locker room jock talk.  

Nerdddddd!

 

Good for him.. I had a seizure reading that

[ControllerOfPlanetX]

Yes..but can he drive a stick?

[Another Fan]

So…. How does this relate to the Bills? 

[Punching Bag]

37 minutes ago, djp14150 said:

What Klein bottle did you fall into on your random walk around the block?

 

I am assuming you are talking about mathematical construct of a Klein bottle not the stuff they sell which is not truly one.  Always loved Möbius strips and imaginary constructs.

 

By age of 6 I read two sets of encyclopedias including Encyclopedia Britannica plus one set of map encyclopedias and was reading Isaac Asimov at 5. I had real problem in elementary school where teachers could not believe I completed work already. Was moved to special school because teachers could not deal with it.  On first day teacher said I was misdiagnosed and specialist gave me am IQ test and I tested as 164.  Tester said that I was reading at 6 levels above level and they were setting up me in special program to correct issues in elementary school.  

 

Children should never be berated for not being at same level as other children.  

They recognized same issue with my sister and she skipped a grade from 1st to 3rd so she did not need to get schooling below her level.

[Mikie2times]

5 hours ago, hondo in seattle said:

 

If you don't want to read the whole thesis, Urschel provided a short summary:

 

"This thesis considers four independent topics within linear algebra: determinantal point processes, extremal problems in spectral graph theory, force-directed layouts, and eigenvalue algorithms. For determinantal point processes (DPPs), we consider the classes of symmetric and signed DPPs, respectively, and in both cases connect the problem of learning the parameters of a DPP to a related matrix recovery problem. Next, we consider two conjectures in spectral graph theory regarding the spread of a graph, and resolve both. For force-directed layouts of graphs, we connect the layout of the boundary of a Tutte spring embedding to trace theorems from the theory of elliptic PDEs, and we provide a rigorous theoretical analysis of the popular Kamada-Kawai objective, proving hardness of approximation and structural results regarding optimal layouts, and providing a polynomial time randomized approximation scheme for low diameter graphs. Finally, we consider the Lanczos method for computing extremal eigenvalues of a symmetric matrix and produce new error estimates for this algorithm."

 

Typical locker room jock talk.  

Can somebody summarize these cliff notes? 

[WotAGuy]

I didn’t ***** myself today, so there’s that. 

[JimBob2232]

And an amazing representative for the Penn State community.

[WhoTom]

1 hour ago, Another Fan said:

So…. How does this relate to the Bills? 

 

6 hours ago, hondo in seattle said:

Urschel grew up in Buffalo

 

[Another Fan]

3 minutes ago, WhoTom said:

 

 

Sounds like it's more Off The Wall discussion then.  Still not Bills related 

 

With all do respect to the OP I could care less 

[ChronicAndKnuckles]

6 hours ago, Big Turk said:

 

Very rare for a person to be gifted with elite traits that allow them to reach the pinnacle of their profession in multiple divergent areas...

Jack Kemp was a pretty successful guy as well in multiple areas. 

[Punching Bag]

34 minutes ago, Another Fan said:

Sounds like it's more Off The Wall discussion then.  Still not Bills related 

 

With all do respect to the OP I could care less 

 

FYI:

The Stadium Wall

Buffalo Bills Community. Discussing the Buffalo Bills and the NFL 24 hours a day since 1998!

 

Report button is available.

[AlCowlingsTaxiService]

7 hours ago, hondo in seattle said:

 

If you don't want to read the whole thesis, Urschel provided a short summary:

 

"This thesis considers four independent topics within linear algebra: determinantal point processes, extremal problems in spectral graph theory, force-directed layouts, and eigenvalue algorithms. For determinantal point processes (DPPs), we consider the classes of symmetric and signed DPPs, respectively, and in both cases connect the problem of learning the parameters of a DPP to a related matrix recovery problem. Next, we consider two conjectures in spectral graph theory regarding the spread of a graph, and resolve both. For force-directed layouts of graphs, we connect the layout of the boundary of a Tutte spring embedding to trace theorems from the theory of elliptic PDEs, and we provide a rigorous theoretical analysis of the popular Kamada-Kawai objective, proving hardness of approximation and structural results regarding optimal layouts, and providing a polynomial time randomized approximation scheme for low diameter graphs. Finally, we consider the Lanczos method for computing extremal eigenvalues of a symmetric matrix and produce new error estimates for this algorithm."

 

Typical locker room jock talk.  

Of course … only an idiot doesn’t know this 

[pennstate10]

Bill O Brien had a funny story about Urschel when he (OB) was head coach at PSU and Urschel was a graduate student, using his final year of eligibility.

 

PSU had a huge whiteboard where they indicated a players practice status.

 

Name in green = full practice

Yellow = limited

Red = no practice

Orange = player was taking a class and would be late

 

One day OBrien sees Urschel's name in purple.  He asks one of the staff-"Why the F is Urschel's name in purple?  What the F does that mean? (OB was pretty foul-mouthed)"

 

Staff says, "oh, that means he's teaching a class".

[Beck Water]

5 hours ago, hondo in seattle said:

Urschel grew up in Buffalo, graduated from Canisius, became a Baltimore Raven, and just got a job as a math professor at MIT.  He wrote his Ph.D. thesis on "Graphs, Principal Minors, and Eigenvalue Problems" and has published over a dozen peer-reviewed papers.  All by the age of 32.  

 

Makes me feel like I haven't tried hard enough.  

 

John Urschel "Player's Tribune" article from 2015

https://www.theplayerstribune.com/articles/why-i-play-football

 

Interview with Urschel from 2020, describing just how hard it actually was to pursue a PhD while playing professional football:

https://www.quantamagazine.org/john-urschel-from-nfl-player-to-mathematician-20200225/

 

 

[Beck Water]

8 minutes ago, pennstate10 said:

Bill O Brien had a funny story about Urschel when he (OB) was head coach at PSU and Urschel was a graduate student, using his final year of eligibility.

 

PSU had a huge whiteboard where they indicated a players practice status.

 

Name in green = full practice

Yellow = limited

Red = no practice

Orange = player was taking a class and would be late

 

One day OBrien sees Urschel's name in purple.  He asks one of the staff-"Why the F is Urschel's name in purple?  What the F does that mean? (OB was pretty foul-mouthed)"

 

Staff says, "oh, that means he's teaching a class".

 

John Harbaugh had a story too. 

 

The NFL encourages players to complete their degrees during the off seasons.  The players need to loop in the HC in case they need some accommodation during OTAs/mandatory minicamp/TC, like a personal day to travel to campus or take tests.  So the coaches are all, well, coached, to support and encourage this.  But they're also busy people so if it's something routine like encouraging a guy to complete his degree, they don't necessarily disengage completely from whatever they're thinking about.

 

So Urschel tapped on Harbaugh's door. 

"Come in!  What can I do for you?" 

"Coach, I wanted to let you know I'm planning to work on my degree this off-season"

"That's great, John, you'll never regret it!  Just let me know how we can help!"

 

Urschel leaves, and Harbaugh goes back to whatever he was doing and then a minute later the lightbulb flashes -

"Wait a minute.  That was Urschel.  What degree is he talking about?"

 

Urschel graduated with his Masters in 2013 before being drafted spring of 2014, and of course once he was focusing, Harbaugh knew that.

[Thurman#1]

10 hours ago, Limeaid said:

 

I was an honors math major in college and read briefly the paper.

https://math.mit.edu/~urschel/publications/thesis.pdf

 

I agree with you I did not work hard enough but I got tired of school taking whatever work I could to pay bills.  At one point I was being chased out of lounges for sleeping there and sleeping in office at my job since I had no housing.

https://dspace.mit.edu/handle/1721.1/140006

 

 

See, that housing problem, you missed out on the easy part. Just become a Baltimore Raven for a few years and earn enough to put you in beds and blankets for decades.

 

Congrats to Urschel, that's great news. He is one of the more fascinating football players ever.

[Beck Water]

4 hours ago, Mikie2times said:

Can somebody summarize these cliff notes? 

 

In short: No.  And I've actually used some linear algebra in my daily life and understand what an Eigenvalue is.

[Billy Claude]

5 hours ago, Beck Water said:

 

In short: No.  And I've actually used some linear algebra in my daily life and understand what an Eigenvalue is.

 

Not surprised.  You can guess who has a data/stat/math background on this board.

 

[freddyjj]

7 hours ago, Beck Water said:

 

In short: No.  And I've actually used some linear algebra in my daily life and understand what an Eigenvalue is.

Isn't that just the value of Rich's 40 yd dash time on NFL combine?

https://www.nfl.com/videos/run-rich-run-year-by-year-results-of-rich-eisen-s-40-yard-dash

 

 

[thenorthremembers]

In risk management/engineering there is a term called "corner case"  where the probability of the hazardous situation occurring is incredibly small.

 

Urschel is a corner case.  As such, no one should feel too bad about not living up to his accomplishments.   That said, well done big fella.

 

 

[C.Biscuit97]

If he went to Joes, he probably be working at Arby’s, living with his “girlfriend” who had kids with 3 other guys, punching windows after he loses in a video game. Canisius = men of excellence 😎

 

he is in the college football hall of fame and retired when he was going to start for the Ravens. So much respect for him. How much can that possibly pay thoigg?