The h Bar

« first day (3303 days earlier) ← previous day next day → last day (1609 days later) »

Aaron Stevens

4:37 AM

@JMac Mine got declined as well

7 hours later…

Novalium Company

11:39 AM

@alarge Ok thx.

why do you even need genetic algorithm in that case. I mean, wouldn't the neural network + the back propagation do the job? I'm just having hard time understanding how neural networks and genetic algorithms blend.

Evolution of Neural Networks using Genetic Algorithm for a 3D car made in Unity

►

I'm watching this video and I understand how the system works but my question is, is it possible to just use only a neural network that will back propagate itself or something?

I honestly can't think of a way how a single neural network could do the job :D

Slereah

11:59 AM

@AbhasKumarSinha I mean, you're integrating wrt $t$

So that function is a constant of $t$, you can just take it out of the integral :p

bolbteppa

12:51 PM

"The first two years of these three years were exceptional with one discovery coming after the other. It started already in the last days of 1970 when Pierre Ramond [14] out of nowhere found the spectrum and the symmetries of the fermionic string.

He did it by following Dirac’s original analysis extending not only the spacetime coordinates but also the γ-matrices to depend on a world sheet variable. This was not only the first piece in the Superstring Theory but also in supersymmetry."

2 hours later…

PM 2Ring

2:28 PM

Oh boy. This answer is so bad it's almost amusing. physics.stackexchange.com/a/515061/123208 And their more recent one is almost incomprehensible physics.stackexchange.com/a/515073/123208 I flagged them as VLQ, I hope that's appropriate.

Aaron Stevens

@PM2Ring Yeah they definitely warrant a down vote. In my experience though the VLQ flags for answers like these are usually declined because the answer still seems to be on topic. I think the only thing to do here is down vote

There is not an option to delete an answer just because it is poorly worded or hard to understand.

JMac

2:49 PM

@PM2Ring Space has stars. My computer has a space bar. Bars serve drinks. Drink in the universe. Universe = universal. If I have 4 universes and add on 0 universes, I have 4 + 0 is 40 times the universes at my disposal. So the question is, do collisions conserve momentum? Yes, yes they do.

PM 2Ring

@AaronStevens Thanks. I figured that VLQ flags are appropriate when it's not possible for a 3rd party to edit the answer into shape.

@JMac Nice work. :D

On SO, I have enough rep to delete vote answers, but that's not something I do lightly. If an answer is merely wrong, I downvote. But if it gives dangerous advice, and the author doesn't respond to comments to fix it (eg, their code is vulnerable to SQL injection attacks), then deletion may be necessary. We also del-vote new answers to old questions if it's a poor duplicate of an existing answer on that question.

JMac

Yeah, I always though VLQ applied to basically nonsense as well. If that answer was actually saying something of substance, and it was wrong, then it should be downvoted. But when you literally can't make sense of what is being said, I don't really see why it should stay.

Aaron Stevens

@JMac I agree. It's just from my experiences that when I flag things like that the flags still get declined.

@JMac I raised a new custom flag on that model of towel wringing answer

JMac

3:07 PM

@AaronStevens Yeah, it seems really weird to me the flags keep getting denied on that one. It literally just says "They model these things in this paper. It seems to agree with experimental results." It seems like a blatant case of link-only to me.

Aaron Stevens

@JMac Yeah I completely agree

Jake Rose

Reading Diracs GR book. In typical fashion he says something with little explanation.

This in regards to the newtonian approximaiton

$g_{\mu \nu}$ are constant in time, $g_{\mu \nu , 0} = 0$. All fine.

But then he says, 'further we must have $g_{m0} = 0$

why??

Aaron Stevens

3:53 PM

What would cause a -5 rep on an answer?

rob

@AaronStevens Generally, five more downvotes than upvotes

Oh, wait, rep.

Aaron Stevens

@rob Yeah. It was on an answer from a month ago

The answer has no down votes

I was just confused since I have never seen that

rob

Perhaps an un-accept and an upvote.

It's an un-accept and an upvote. physics.stackexchange.com/users/179151/…

Aaron Stevens

@rob Ah ok. Yeah don't know why I didn't consider looking there haha

Thanks

rob

@AaronStevens I've just noticed from your profile that we're neighbors.

Aaron Stevens

4:05 PM

@rob Where are you?

rob

@AaronStevens Nashville area.

Aaron Stevens

@rob Oh, well how exciting

I might be leaving soon, depending on where I get a job after I graduate

Although I do have one postdoc opportunity at Vanderbilt

rob

@AaronStevens It's usually a good idea to postdoc somewhere other than your grad school.

Aaron Stevens

@rob True, but I wouldn't be staying in the same department

rob

@AaronStevens That helps.

Aaron Stevens

4:10 PM

The opportunity is in a Cell and Developmental Biology group over in the medical center

Who would need some biophysics expertise

rob

@AaronStevens That actually sounds pretty exciting.

Aaron Stevens

@rob Yeah that is what I am thinking. The thing I am a little worried about is that it is a biology group. The attitude was kind of like "Hey we have all of these things that would greatly benefit from modeling/physics perspectives"

As opposed to other postdocs I have been looking at where they already have the modeling/physics established

rob

@AaronStevens The way to address that concern (which is a good one) is to find some folks who have made that kind of discipline-crossing transition and ask them about it.

There may be earlier graduates from your current group that you could talk with.

Aaron Stevens

@rob Yeah that's true. I also haven't had the chance to talk to my adviser about it yet. I just met with the head of this group on Friday

But not having to move would be a bonus haha

I have been in Nashville for over 8 years now, so it really feels like home

rob

It's a nice town. Friendly folks.

Aaron Stevens

4:22 PM

I just need them to hurry up and finish the 440 construction

rob

No kidding.

Aaron Stevens

They don't need to be done until August 2020 though

DanielSank

4:37 PM

Hi, everybody.

@AaronStevens Jumping into this way too late, but I have opinions.

@rob Agreed.

Aaron Stevens

@DanielSank Go for it

Aaron Stevens

4:51 PM

Nevermind

bolbteppa

5:04 PM

@JakeRose sections 1 and 2 here explain why being static leads to that holding:

Deriving the Schwarzschild solution

The Schwarzschild solution describes spacetime in the vicinity of a non-rotating massive spherically-symmetric object. Of the solutions to the Einstein field equations, it is considered by some to be one of the simplest and most useful. == Assumptions and notation == Working in a coordinate chart with coordinates ( r , θ , ϕ , t ) {\displaystyle \left(r,\theta ,\phi ,t\right)} labelled 1 to 4 respectively, we begin with...

2 hours later…

jacob1729

6:55 PM

Q: Extending Coordinate Transformation in Relativity

I have been trying to teach myself general relativity, and I come from a background in mathematics, but I am having a difficult time understanding a particular math fact my book is using. The author is trying to develop the field equations for free space, and starts by proving every point $P$ has...

Can someone explain how the claim in this question squares with gravitational waves existing?

(Author claims that in vacuum, there is a coordinate transformation to being globally Minkowski space time. I thought the whole point of gravitational waves was that they were not coordinate artefacts?)

DanielSank

7:32 PM

@AaronStevens If you can find a bio lab that's already doing well independently, then joining and adding your expertise could be awesome.

But don't expect to fix a broken situation.

Cross-discipline work can be outstandingly productive.

2 hours later…

Aaron Stevens

9:29 PM

@DanielSank I agree. So far the options I have been considering all have pretty good work being done.

DanielSank

9:55 PM

tjat