Search found 201 matches

by necroforest
Fri Jan 17, 2014 12:08 pm UTC
Forum: Individual XKCD Comic Threads
Topic: 1318: "Actually"
Replies: 76
Views: 16692

Re: 1318: "Actually"

IIRC, EGM96 defines the gravitational potential (or, deviation from an ideal gravitational potential), but not the actual shape. The shape is typically modeled as an ellipsoid (WGS84) + local topography (which is usually a sampled grid of 'height-above-ellipsoid' values)
by necroforest
Mon Jan 03, 2011 2:01 pm UTC
Forum: Computer Science
Topic: Compiling languages into C
Replies: 9
Views: 3303

Re: Compiling languages into C

Check out Lush - it's a scientific-oriented Lisp dialect that compiles into C.
by necroforest
Mon Jan 03, 2011 1:39 pm UTC
Forum: Computer Science
Topic: Grad school for comp sci?
Replies: 14
Views: 5559

Re: Grad school for comp sci?

You might also consider working for an employeer who will pay for you to get a part-time Master's degree. You get "real world" experience on top of free school. If you have the time, it would probably be in your best interests to at least pick up a minor in CS.
by necroforest
Mon Jan 03, 2011 2:41 am UTC
Forum: Computer Science
Topic: Feels type insecure? (informal discussion of opinions)
Replies: 15
Views: 3120

Re: Feels type insecure? (informal discussion of opinions)

"int" instead of "var" ... yeah, there goes my day, better to crash at runtime. Clearly not a Java developer :) Map<String, List<Foo<Long,Map<String,String>>>> myMap = new HashMap<String, List<Foo<Long,Map<String,String>>>>(); (ok, that's a little exaggerated.... usually )
by necroforest
Wed Nov 10, 2010 3:07 am UTC
Forum: Science
Topic: Modeling physics problems
Replies: 7
Views: 1088

Re: Modeling physics problems

BlackSails wrote:How about tackling some open question about the numerical behavior of the ising model?


If it's an open question it's probably above my abilities :) Do you have a paper reference I could read about this?
by necroforest
Wed Nov 10, 2010 2:35 am UTC
Forum: Science
Topic: Modeling physics problems
Replies: 7
Views: 1088

Re: Modeling physics problems

the_tastiest_sandwich wrote:You could approach the problem from a purely mathematical point of view ("here's a differential equation in physics, the derivation of which I couldn't follow, and here are some numerical simulations!").


That's essentially what i'm aiming for, and is basically what the project is.
by necroforest
Wed Nov 10, 2010 2:32 am UTC
Forum: Science
Topic: Modeling physics problems
Replies: 7
Views: 1088

Re: Modeling physics problems

BlackSails wrote:What level class is this?


Graduate level CS/applied math course.
by necroforest
Tue Nov 09, 2010 10:59 pm UTC
Forum: Science
Topic: Modeling physics problems
Replies: 7
Views: 1088

Modeling physics problems

I'm taking a class on "modeling & simulation of complex systems". We have to do a semester project where we pick a problem and model it with whatever tools we want. So, what I want to do is find a physics problem (preferably in the area of electrodynamics, particle physics, or GR) to m...
by necroforest
Fri Jul 23, 2010 1:11 pm UTC
Forum: Mathematics
Topic: How do we know i is well defined?
Replies: 16
Views: 2869

Re: How do we know i is well defined?

There isn't commutativity in the quaternions (hence the need to call them a ring, not a group)


Groups don't have to be commutative...
by necroforest
Wed Jun 23, 2010 9:14 pm UTC
Forum: Computer Science
Topic: Power with carry.
Replies: 4
Views: 1202

Re: Power with carry.

are you asking about how to compute [imath]a^x mod b[/imath] ? if so, you can just do successive squaring, and doing the mod at each step to keep the results reasonably bounded.
by necroforest
Wed Jun 09, 2010 5:18 am UTC
Forum: Mathematics
Topic: Self Teaching PDE's?
Replies: 5
Views: 1461

Re: Self Teaching PDE's?

http://www.amazon.com/Differential-Equa ... 598&sr=8-1

This is pretty good (although i'm no PDE guru), and has the advantage of being pretty cheap for a math book.
by necroforest
Thu May 27, 2010 1:53 pm UTC
Forum: Mathematics
Topic: Returning to School
Replies: 9
Views: 1757

Re: Returning to School

Doesn't your company have a tuition reimbursement program for masters degrees?
by necroforest
Thu May 27, 2010 4:47 am UTC
Forum: Mathematics
Topic: Returning to School
Replies: 9
Views: 1757

Re: Returning to School

You could get a masters first, which would give you research experience (thesis) and people to recommend you.

I think if you pick your programs right, you can get out of a lot of classes in a doctoral program if you already have a masters.
by necroforest
Fri May 21, 2010 1:13 pm UTC
Forum: Computer Science
Topic: Help with Neural Networks
Replies: 12
Views: 4009

Re: Help with Neural Networks

basically, the weights of a neuron forms a line (or hyperplane) through your input space, and the activation function tells you which side of the line you're on (and how far from the line, if you're using a sigmoid function). if you have a bunch of neurons organized in layers, the first layer chops ...
by necroforest
Sun Feb 28, 2010 11:12 pm UTC
Forum: Coding
Topic: Is Java that bad?
Replies: 114
Views: 19309

Re: Is Java that bad?

My views of Java (as a professional Java developer... oh god, did I just say that?) Pros: - Very comprehensive standard library - Lots and lots and lots of add-on libraries - Cross-platform Cons/things I just don't like: ⋅ Lack of destructors for classes and lack of "deterministic&quo...
by necroforest
Wed Jan 27, 2010 4:21 am UTC
Forum: Mathematics
Topic: Everyday randomness
Replies: 85
Views: 9617

Re: Everyday randomness

That doesn't change the fact that Shor's algorithm doesn't apply to AES. AES is not based on the (conjectured) hardness of factoring. Also, from Wikipedia: http://en.wikipedia.org/wiki/Key_size#Effect_of_quantum_computing_attacks_on_key_strength Mainstream symmetric ciphers (such as AES or Twofish) ...
by necroforest
Tue Jan 26, 2010 5:31 am UTC
Forum: Mathematics
Topic: Everyday randomness
Replies: 85
Views: 9617

Re: Everyday randomness

Shor's algorithm doesn't apply to AES. AFAIK, quantum computing doesn't break AES like it breaks RSA, although I believe it gives you a quadratic speedup (Grovers algorithm I think?). This means that you'd need to double the key size for the same protection (i.e, AES-256 is only as strong as AES-128...
by necroforest
Sun Dec 20, 2009 2:00 pm UTC
Forum: Coding
Topic: Scheme IDE
Replies: 7
Views: 2483

Re: Scheme IDE

I don't know about Scheme, but for Common Lisp the standard editor seems to be SLIME:
http://common-lisp.net/project/slime/
by necroforest
Sat Dec 12, 2009 5:40 pm UTC
Forum: Coding
Topic: Perl, Ruby, or Common Lisp?
Replies: 11
Views: 2500

Re: Perl, Ruby, or Common Lisp?

Out of that list, I'd go with Perl - you know scheme so that's pretty close to Lisp. I use Perl at work and on some grad school projects and love it.

Outside of the list, I'd go with Haskell - it's pretty different from everything on your list.
by necroforest
Sat Oct 31, 2009 3:08 pm UTC
Forum: Computer Science
Topic: The Calculus
Replies: 4
Views: 1713

Re: The Calculus

backpropagation algorithm, Akra-Bazzi theorem
by necroforest
Sun Oct 25, 2009 11:12 pm UTC
Forum: Computer Science
Topic: Neural Network applications
Replies: 7
Views: 1886

Neural Network applications

I'm currently taking a grad-level class in Neural Networks, and we have to do some type of project (i.e, write a neural network that does x, where x could be facial recognition/feature extraction/stock market prediction/etc). Anyways, I was wondering if anyone in here who has experience with neural ...
by necroforest
Sun Oct 04, 2009 12:11 pm UTC
Forum: Mathematics
Topic: Set theory question
Replies: 17
Views: 1892

Re: Set theory question

Didn't see that. I was rather surprised that they didn't say "not necessary disjoint". This was my lower level "intro to sets / proofs" class, so everything was a bit simplified. Of course, you could just as easily have defined ƒ as f=\begin{cases}g(\frac{n+1}{2}) & ...
by necroforest
Sun Oct 04, 2009 2:23 am UTC
Forum: Mathematics
Topic: Set theory question
Replies: 17
Views: 1892

Re: Set theory question

A and B are disjoint, so [imath]A\cap B = \emptyset[/imath].
by necroforest
Sat Oct 03, 2009 11:32 pm UTC
Forum: Mathematics
Topic: Set theory question
Replies: 17
Views: 1892

Re: Set theory question

Um, here's an old homework I dug up: Let A,B be countable, disjoint sets. Prove that A\cup B is also countable. To prove this, you need to show there's a bijective function f:\mathbb{N} \rightarrow (A\cup B) The way I did it in the proof (which is easy but I don't feel like re-typing the who...
by necroforest
Sat Oct 03, 2009 1:46 am UTC
Forum: Mathematics
Topic: Set theory question
Replies: 17
Views: 1892

Re: Set theory question

You can also look at functions as being defined by sets. For example, the function f(x)=x^2 could be defined as the infinite set of input/output pairs: f = \{ (0,0), (1,1), (2,4), (1.5,2.25) \} (and the ordered pairs can be described as sets themselves, i.e &#...
by necroforest
Wed Jul 08, 2009 10:38 pm UTC
Forum: Mathematics
Topic: Optimization Problem
Replies: 6
Views: 928

Re: Optimization Problem

dunderthunder wrote:that uses the minimum amount of steel


Build it out of titanium.
by necroforest
Thu Jun 11, 2009 10:41 pm UTC
Forum: Mathematics
Topic: Number theory / CS problem
Replies: 9
Views: 1011

Re: Number theory / CS problem

PRIMES is in P , and there are even more efficient (in practice) probabilistic primality tests. I think you meant to say that this is at least as hard as prime factorization, which is believed to be hard. The fastest deterministic primality testing algorithm is something like \Theta(n^{12})...
by necroforest
Mon May 25, 2009 3:29 am UTC
Forum: Computer Science
Topic: Introduction to cryptography
Replies: 9
Views: 2079

Re: Introduction to cryptography

Stinson (mentioned above) has some cryptanalysis in it (including linear and differential), but its not dedicated to cryptanalysis
by necroforest
Mon May 25, 2009 1:19 am UTC
Forum: Science
Topic: Blocking a Nuclear Missile
Replies: 32
Views: 3773

Re: Blocking a Nuclear Missile

This might work, depending on the details of the rocket (i.e, if it's a liquid fueled rocket where the fuel is controlled by some computer) (1) Wait for ICBM to be over an ocean/large body of water (2) Fire your missile at it. The missile has some EMP device that is triggered when in proximity to th...
by necroforest
Thu May 21, 2009 12:57 am UTC
Forum: Science
Topic: Non-science undergrad in science grad school
Replies: 10
Views: 1924

Non-science undergrad in science grad school

I was going to post this in "school" but that seems to be more about high school and undergrad than graduate school, so I figured this would be a better place. Anyways, here's my question: I have an undergrad degree in computer science (currently working as a software engineer), and am pre...
by necroforest
Wed May 13, 2009 2:02 am UTC
Forum: Computer Science
Topic: Computer Science Colleges - A Mega-Merg'd Thread
Replies: 52
Views: 5989

Re: How to tell a good Computer Science Department

If you already have some decent programming knowledge, I would consider doing engineering/science/math and maybe minoring in CS... it seems like it's easy for an engineer to get a CS job, where it's not easy for a CS to get an engineering job. I'm currently a software engineer and 3 of the other eng...
by necroforest
Mon May 11, 2009 10:55 pm UTC
Forum: Science
Topic: You might be a physics major if...
Replies: 540
Views: 64817

Re: You might be a physics major if...

YMBAPM if you describe your favorite color by its wavelength. What if you favorite color isn't describable as a pure wavelength? Then you'll have to show your physics major-osity in other ways. 520 nm. YMBAPM if you describe your favorite color by its spectrum. YMBAPM if you describe your favourite...
by necroforest
Mon May 11, 2009 10:37 pm UTC
Forum: Science
Topic: You might be a physics major if...
Replies: 540
Views: 64817

Re: You might be a physics major if...

Nlelith wrote:
Meteorswarm wrote:
Nlelith wrote:YMBAPM if you describe your favorite color by its wavelength.


What if you favorite color isn't describable as a pure wavelength?

Then you'll have to show your physics major-osity in other ways. 520 nm.


YMBAPM if you describe your favorite color by its spectrum.
by necroforest
Mon May 11, 2009 9:54 pm UTC
Forum: Science
Topic: You might be a physics major if...
Replies: 540
Views: 64817

Re: You might be a physics major if...

Ugh. About 75% of these apply to me... Maybe I should have been a physics major.

/Comp Sci major
by necroforest
Mon May 11, 2009 1:26 am UTC
Forum: Computer Science
Topic: Introduction to cryptography
Replies: 9
Views: 2079

Re: Introduction to cryptography

Here's what I used in my crypto class:

Stinson - Cryptography: Theory and Practice

http://www.amazon.com/Cryptography-Prac ... 137&sr=8-1
by necroforest
Sun May 10, 2009 4:16 pm UTC
Forum: Mathematics
Topic: Help! I'm trying to get a 14 year old interested in Maths.
Replies: 52
Views: 7124

Re: Help! I'm trying to get a 14 year old interested in Maths.

Does phi rhyme with "rye" or "pee" ?
by necroforest
Mon Mar 02, 2009 5:06 am UTC
Forum: Individual XKCD Comic Threads
Topic: 0550: "Density"
Replies: 184
Views: 66625

Re: "Density" Discussion

Next time, listen to Advice Dog instead of Courage Wolf.
by necroforest
Thu Feb 05, 2009 2:08 am UTC
Forum: Mathematics
Topic: Something I found while doing DFTs...
Replies: 6
Views: 953

Something I found while doing DFTs...

I have homework to find close forms for DFTs of various functions. Anyways, I was simplifying some ugly expressions and had this train of thought: \forall x \in R, e^{ix}=1 Proof: e^{ix} = e^{2\pi i x/2\pi}=(e^{2\pi i})^{x/2\pi}=(1)^{x/2\pi}=1 You can prove this result for any comple...
by necroforest
Tue Nov 04, 2008 4:50 am UTC
Forum: Fictional Science
Topic: What is the mass of a feather?
Replies: 229
Views: 321787

Re: What is the mass of a feather?

Uh, you're all really wrong. I would have expected better of you. Now, since feathers are used for flying, they go up and not down in a gravitational field [the earth]. This means that they have negative mass. So the correct way to find the mass is to hold a scale upside down and let a feather go un...
by necroforest
Fri Oct 31, 2008 9:24 pm UTC
Forum: Computer Science
Topic: Research for MS
Replies: 8
Views: 2067

Re: Research for MS

I'd recommend doing a Straight-edge-and-compass proof that P=NP

Go to advanced search