For Sample Use Only: Fill & Download for Free

I note that there are no responses to this question from people who were killed by the mugger. In physics, we would say that the sample of people who answered reflects a strong observational bias. Each response from someone who was mugged and survived shows what worked at least once. That same answer may have failed 99% of the time, and we would not know it.

There are various artificial, esoteric programming languages which use all kinds of peculiar character sets. For example, Whitespace uses only white spaces, and Brainfuck code looks like this:++++++++[>++++[>++>+++>+++>+<<<<-]>+>+>->>+[<]<-]>>.>---.+++++++..+++.>>.<-.<.+++.------.--------.>>+.>++. But among actual programming languages meant for use by actual programmers, the one that comes to mind is APL. It deliberately relied substantially on special characters, and this is likely one of the reasons it never became mainstream.Code sample (from the Wikipedia entry):(~R∊R∘.×R)/R←1↓ιR 

For a quick simple explanation:In both gradient descent (GD) and stochastic gradient descent (SGD), you update a set of parameters in an iterative manner to minimize an error function.While in GD, you have to run through ALL the samples in your training set to do a single update for a parameter in a particular iteration, in SGD, on the other hand, you use ONLY ONE or SUBSET of training sample from your training set to do the update for a parameter in a particular iteration. If you use SUBSET, it is called Minibatch Stochastic gradient Descent.Thus, if the number of training samples are large, in fact very large, then using gradient descent may take too long because in every iteration when you are updating the values of the parameters, you are running through the complete training set. On the other hand, using SGD will be faster because you use only one training sample and it starts improving itself right away from the first sample.SGD often converges much faster compared to GD but the error function is not as well minimized as in the case of GD. Often in most cases, the close approximation that you get in SGD for the parameter values are enough because they reach the optimal values and keep oscillating there.If you need an example of this with a practical case, check Andrew NG's notes here where he clearly shows you the steps involved in both the cases. http://cs229.stanford.edu/notes/cs229-notes1.pdf