Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
Closed 8 years ago.
Questions concerning problems with code you've written must describe the specific problem — and include valid code to reproduce it — in the question itself. See SSCCE.org for guidance.
This question does not appear to be about a specific programming problem, a software algorithm, or software tools primarily used by programmers. If you believe the question would be on-topic on another Stack Exchange site, you can leave a comment to explain where the question may be able to be answered.
Improve this question
Could someone please explain what large margin optimization is, in a machine learning context? Everything I've found is extremely complicated and I don't know where to start.
Thanks in advance.
In classification, the largest margin problem is simply a search for the separating boundary (hyperplane in most cases) which maximizes the margin aroud it (minimum distance to the objects of each class).
In the simple case of two dimensional data, you can think of it like a search for such line, that correctly separates elements of one class from the other one, and in the same time maximizes the sum of distances to the closest points from both classes. Following image from wikipedia shows such separating line found using Support Vector Machine:
This geometrical concept is very important, as it makes solution unique - if we would simply search for a line that separates our data we would have infinitely many such solutions, and would have a problem to choose a particular one. Largest margin concept shows us exactly which line we want, and as a result, optimization process performed in this problem is generally repeatable (and as numerous experiments have shown - very effective).
Related
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 2 years ago.
Improve this question
Can anyone please help me with this , How is the DP's Iterative approach different from the Recursive Approach.
Dynamic Programming and Recursion aren't necessarily opposites. What you're thinking is Memoization vs. Dynamic Programming.
Dynamic Programming is the approach to a problem that reduces duplicate computations as much as possible. This usually means taking a bottom-up approach - i.e. you calculate answers to smaller scale problems first and then use those answers to calculate higher order problems. Iterative approaches are usually used for Dynamic programming since it seems natural (although you can do it recursively too).
Memoization is the top-down approach to a problem and is usually done through recursion because it is more natural. In this case, you start with a higher order problem and make recursive calls for lower order problems in order to solve it.
In both cases you use a data-structure to store values computed so far.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
Closed 3 years ago.
Edit the question to include desired behavior, a specific problem or error, and the shortest code necessary to reproduce the problem. This will help others answer the question.
This question does not appear to be about a specific programming problem, a software algorithm, or software tools primarily used by programmers. If you believe the question would be on-topic on another Stack Exchange site, you can leave a comment to explain where the question may be able to be answered.
Improve this question
This is the question
Select three random companies, and issue the whois and traceroute (tracert in Windows) commands for each one. Tracert is available from a command prompt. To use whois, you will need to search for an online tool. Then write a short paragraph about each utility outlining the kinds of information available from each. Copy & Paste screen shots for each utility and each company to back up the reported findings.
Assuming i am a noob. I would be glad if someone would outline how to tackle this question for my homework assignment.
This is what I make of this question
Tracert
Read this article for information on trace routes
Company 1 - BBC
Run a tracert on bbc.co.uk
C:\>tracert bbc.co.uk
Using the information from the tracert documentation will help you document your findings. Then follow the same methodology for the next two companies.
Who is
There is a webisite Whois Lookup
Just put in bbc.co.uk or whatever.
The first paragraph here outlines the information you can get from it. Also looking up a domain will show you what information this will provide.
The information on the websites will help with:
Then write a short paragraph about each utility outlining the kinds of information available from each.
You will then need to take screenshots.
The question is very straight forward, add a sprinkling of common sense and you have yourself an answer.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 7 years ago.
Improve this question
Can anyone explain how the functional programming model differs from the procedural or object orientated models.
I cannot conclude a good answer myself.
in my opinion FP is about pure functions (that is functions in a mathematical sense) - which implies referential transparency and, if you continue the thouhgt, immutable data.
This is the biggest difference I see: you don't mutate data - and most other aspects either directly follow from this or from cool type-systems (which are not necessary for a language to be called functional) and the academic nature.
But of course there is far more to it and you can read papers, complete books or just wikipedia about it.
please note that you can dispute the pure property and then things get a lot more fuzzy ... which should not surprise you, as most functional languages in wide to allow for mutation (Clojure, Scala, F#, Ocaml, ...) and there are not many pure ones.
In this case the biggest difference might be the way you abstract things with higher-order-functions (at least functions should be first class citizens - meaning you can pass them around and have them as values).
But overall this question is really opinionated and will very likely be closed as to broad or something - maybe you should ask for details instead of the big picture
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 8 years ago.
Improve this question
I have done some searching around the Internet and SO looking for an introduction or analysis of what makes data.table so fast, but I've only found a lot of (very helpful) manuals, no breakdown of what goes into the programming. (I am more or less completely floored that I can't locate a published paper for data.table, not even something from JStatSoft.)
I've had an algorithms class so I know about sorts and linked lists and binary trees and such, but I don't want to make any amateur guesses (especially when I go to explain to academic people why it's a good idea to use it). Can anyone offer a short, topical summary with references? This question references a slide presentation which is cool, but the info comes in pieces (and even the documentation for, say, setkey() doesn't cite a data.table reference, but goes to Wikipedia).
What I am looking for is something that is both not the source code and not a list of Wikipedia topics, but an ideally "official", sourced answer (thus making it canonical, which could help a lot with all the questions orbiting around this topic).
(It would be great if there was a technical paper out there I could cite for this (the citation() for data.table is just the manual, but of course it's not directly relevant to the question as far as SO is concerned.)
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 9 years ago.
Improve this question
The idea that one gives some of his credits to get an answer for a question seems simple enough but when applied to a whole network of millions of users each with varying reputations it seems like some rigorous checks and balances have to be applied. What is the math framework on which this stuff is based?
There's no math behind the bounty system. It's just rules that are described in the FAQ - https://stackoverflow.com/faq#bounty
The mathematical basis of this system would be simple addition and subtraction. Why the range was chosen is probably more because of economics rather than math. Supply and demand ... The demand is high and the supply has been found to be low, therefore the "price" will be relatively high. 50 to 500 reputation points is the reward set which may be construed as "relatively high" to enough consumers(answerers) to make it worth their time and effort to "purchase" the reputation points with a good answer. The bounty could also be construed as a marketing ploy to help market the product, thereby achieving the sale; which again is the sought after good answer. This bounty system is therefore a great example of market economics; and in the past, it was often noted that one's "reputations" is worth more than money.