Difference between revisions of "Machine learning"
|Line 58:||Line 58:|
Revision as of 16:28, 17 January 2017
Machine Learning is a type of AI that is best described by Tom Mitchell as: "A computer program is said to learn from experience (E) with respect to some class of tasks (T) and performance measure (P), if its performance at tasks in T, as measured by P, improves with experience E." 
Types of Machine Learning
Machine Learning is split into 2 categories; Supervised and Unsupervised Learning.
Supervised learning can be described as a type of machine learning where we already know what the output should look like. If we know the relationship of X and Y (as an example), this gives us a basis for knowing what the output will look like.
Categories of Supervised Learning
If we are given a selection of prices of cars relative to their model year, we can find out where a specified model year may result in a specific price.
In order to use a regression problem, we must have a continuous output. Price being the case here.
If we are given a selection of variables such as Age and Height, we may wish to group these variables to find out the chance of someone who is perhaps 16 years old, being a certain height or not.
In order to do this, we must classify or group inputs into categories in order to make an informed prediction.
Unsupervised learning is used where we have little or no idea what the results may look like.
With unsupervised learning, data can be clumped or clustered into groups based on the variables in the data. By doing so, we can create new structure from a data set.
It can also be used in the reverse using what is known as "non-clustering" to find specific data from a larger array. An example of non-clustering is the "cocktail party algorithm" which allows you to find structure in a chaotic environment.
A well known example of unsupervised learning is Google news.
Machine Learning and Transhumanism
Differences between Machine learning and AI
An article on Wired UK mentions that: "AI is a branch of computer science attempting to build machines capable of intelligent behaviour, while Stanford University defines machine learning as “the science of getting computers to act without being explicitly programmed”. You need AI researchers to build the smart machines, but you need machine learning experts to make them truly intelligent."
It also states:
"We spoke to Intel’s Nidhi Chappell, head of machine learning to clear this up.
“AI is basically the intelligence – how we make machines intelligent, while machine learning is the implementation of the compute methods that support it. The way I think of it is: AI is the science and machine learning is the algorithms that make the machines smarter.
“So the enabler for AI is machine learning,” she added."