For the past several months I've been interested in some Machine Learning and I can't help but notice that nomatter what kind of learning algorithm I'm using - at its core lies a simple polynomial or logistic regression. Today I went back to the most basic definition that I could find:

Machine learning, a branch of artificial intelligence, concerns the construction and study of systems that can learn from data.

The fact that we're creating mathematical hypotheses based on data ALWAYS means that you're doing some kind of statistical regression ?

If I'm wrong, can you please provide an example where this doesn't hold true.


closed as unclear what you're asking by Patrick87 Feb 24 '14 at 16:00

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 3
    $\begingroup$ I recommend looking at computational learning theory (the theory branch of ML), a lot of the algorithms and techniques there don't look like statistics (although in the vaguest sense of the word, they are achieving similar goals). $\endgroup$ – Artem Kaznatcheev Feb 24 '14 at 9:21
  • 1
    $\begingroup$ Algorithms learning decision trees look different. Genetic algorithms look different. Even neural networks look different. Among the (strong) common methods I know of, only support vector machines use something similar to linear regression $\endgroup$ – Yuval Filmus Feb 24 '14 at 15:24
  • $\begingroup$ Hello, and welcome to Computer Science StackExchange. This question has a lot of potential but, as written, is not currently a great fit for this (or other) SE network sites. This question advances a hypothesis and asks for confirmation; if correct, the only correct answer is "yes", and if wrong, there may be too many good answers. Please feel free to visit the help center, our Meta site, or our chat room to discuss how to improve this question. Once the question has been edited, please flag it so that a moderator can reopen. Thanks again for contributing to the site, and again, welcome! $\endgroup$ – Patrick87 Feb 24 '14 at 16:05

Browse other questions tagged or ask your own question.