Can Wolfram Alpha do n-d integrals symbolically?

  • Thread starter Thread starter benorin
  • Start date Start date
  • Tags Tags
    Integrals
AI Thread Summary
The discussion centers around the capabilities of Wolfram|Alpha in handling n-dimensional integrals, specifically Selberg integrals. The term "n-d" is clarified to mean "n-dimensional," which some participants found unclear. It is noted that Wolfram|Alpha is primarily designed for simpler problems, which may limit its ability to process complex integrals due to character input restrictions. The character limit for input has evolved from about 80 to approximately 125 characters, impacting the complexity of problems that can be solved. Participants suggest that translating complex mathematical notation into a format understandable by Wolfram|Alpha can be challenging and may require breaking down problems into simpler components. There is an invitation for users to present simpler examples of Selberg integrals to explore potential solutions and methods for effectively using Wolfram|Alpha for more complex queries.
benorin
Science Advisor
Insights Author
Messages
1,442
Reaction score
191
The Q is all in the title. I’ve been working on Selberg integrals, curious if Alpha can help me?
 
Physics news on Phys.org
benorin said:
The Q is all in the title.
The title is "Can Wolfram|Alpha do n-d integrals symbolically?"
I didn't know what n-d was supposed to mean, so I looked up Selberg Integral on Wikipedia, which didn't help.

I finally concluded that possibly you meant "n-dimensional" integrals. It would have been helpful to spell this out a bit more, if that's what you actually meant.
 
Mark44 said:
The title is "Can Wolfram|Alpha do n-d integrals symbolically?"
I didn't know what n-d was supposed to mean, so I looked up Selberg Integral on Wikipedia, which didn't help.

I finally concluded that possibly you meant "n-dimensional" integrals. It would have been helpful to spell this out a bit more, if that's what you actually meant.
n-dimensional yes, sorry I thought "n-d" was standard. These are the Selberg Integrals of which I speak:

unit hypercube integrals.png
 
I think WolframAlpha was originally intended to do "small" problems more easily for students than something like Wolfram Mathematica.

Part of the design decisions that went into WolframAlpha appear to have included a line length limit (at least for the free online version, I can't speak about the paid "pro" version of WolframAlpha), you can't enter a problem that takes 250 characters to input or that needs 250 characters in some intermediate or final step in doing the problem. For lots of years I think I remember that limit appeared to be about 80 characters, but it seems that the limit was increased to about 125 characters years ago. I am guessing that limit may be one of the issues that you may face in trying to do the problems you want to solve.

Another issue will likely be trying to translate the abstract notation that makes perfect sense to you and everyone else who have been working for years on the kind of problems you are interested in into a notation that WolframAlpha is able to understand. For some problems that process is fairly straight forward, sometimes it requires a few tricks. Sometimes the line length limit can be avoided by breaking the problem down into a sequence of simpler problems, giving each one to WolframAlpha and finally reassembling the individual results into a complete solution. And for the rest of all the problems that might be posed my response is usually "I don't know of a way to format that so that WolframAlpha will understand and I suggest that you get a 'grad student 2.0' to solve that for you."

If you can show one or a few of the simplest shortest Selberg Integrals that you are interested in then I will take a moment and see if I can find a way to coax WolframAlpha into doing those for you. Then you might include a couple of intermediate sized problems and finally an example or two of the kind/size problems that you really want to do. Maybe we can find a way to get it to do some of your problems and you can see the result of the ad hoc methods I have used to coax WolframAlpha into cooperating and if that works then maybe you can guess how you might try similar things and see if it can solve some of your problems you are really interested in.
 
Back
Top