Part of Our Problem

MeanGreenTexan · January 10, 2018

I think gomeangreen.com leads the entire internet in Drew Brees, Russell Wilson, & Doug Flutie references.

MeanGreenHoops · January 10, 2018

25 minutes ago, MeanGreenTexan said:

I think gomeangreen.com leads the entire internet in Drew Brees, Russell Wilson, & Doug Flutie references.

We'd be Oklahoma, if only we were a few inches taller.

Cerebus · January 10, 2018

On 1/9/2018 at 10:32 AM, Aldo said:

It's ok to say Vanderbilt > North Texas. Really is.

If you are talking about academics then it is just as valid to say North Texas > Vanderbilt. Really is.

Depends entirely on the department you are talking about. NT has some mediocre programs, some that are top in the state, and a handful that are as good or better than any public university in the nation.

Now are some of our best programs likely to be the ones that the average student athlete would be interested in? Maybe, maybe not.

GMG24 · January 11, 2018

On 1/9/2018 at 11:48 AM, GrandGreen said:

Not an expert on Vandy. However, my guess would be athletes are steered to classes that keeps them eligibility not necessarily ones that are going to prepare them for some great career once out of college. All fb division schools do this to some degree, but at so called elite academic schools, it is escalated.

So that degree from Vanderbilt may not be so great when an employer actually looks at the academics behind it.

Kid I coached in bball went up there and double majored while being an all sec hooper. Now on Nicks practice squad I believe

UNT90 · January 13, 2018

On ‎1‎/‎9‎/‎2018 at 10:28 PM, UNTLifer said:

Wild ass speculation or factual, unbiased statement? I go with wild ass speculation.

rah rah rah, sis boom bah.

UNT90 · January 13, 2018

On ‎1‎/‎9‎/‎2018 at 7:45 PM, Harry said:

I disagree. Look at the draft every year. There are high level picks from small schools every year. If they have the ability size and speed the teams will find them anymore.

Zach Orr was a pro bowler year 3.

Size may cost you to drop from a 3rd to a 5th rounder at other schools. At UNT, it drops you out of the draft completely. Because that is UNT's reputation at work. With any job done by humans, it will be influenced by human nature. Human nature tells one that when a school perpetually sucks at football and hasn't produced a draft pick in 15 some odd years, its not going to produce one this year. Human perceptions are VERY hard to change.

If you believe anything else, you are naïve.

And no, computers don't solve anything, because they are programmed in what to look for by humans.

Cerebus · January 13, 2018

2 hours ago, UNT90 said:

And no, computers don't solve anything, because they are programmed in what to look for by humans.

Actually, no not really. Not anymore. Not with the newest AIs. That is what machine learning is all about.

Look, lets take reading handwriting as an example. The first thing you had to do was teach a machine how to read an single letter, for example A. Now the problem with that is how do you describe what an A looks like. It's hard enough to do that with typescript, how in the world can you describe every single A in any handwriting ever? This is an example of as task that is hard for a general AI like a human being, think about it, it takes years to teach a person to read. For a specific AI like a computer program, it is an incredibly hard task to program.

What you do now is build a neural network and then train it. Instead of trying to describe what an A looks like, you show it an A. And then you show it a letter that is not an A. Then you do that with different samples millions if not billions of times. The AI "learns" what as A is and what isn't an A.

Reading an A or even reading handwriting is a simple use case for machine learning. A more complex use case is cancer diagnosis and treatment. Studies already show that several AIs are better than doctors at diagnosing certain cancers. Hundreds of millions of dollars are being used to develop AIs for treatment, with promising results so far.

Now, is anyone spending money to build an AI for NFL draft picks? Not that I know of, but it will be done sooner or later. That's the beauty of computing, everything expensive becomes cheap sooner or later.