Difference between ML and AI
No description yet. This curated deck is ready for daily study.
Card preview (9)
What are the two main applications of Artificial Intelligence (AI) and Machine Learning (ML)
Classification and Regression
What is supervised learning
When the Ml algorithm learns a function that maps an input to an output based on examples of input-output pairs. e.g learns what is a dog from pictures that are labelled as dogs
What is unsupervised learning
When the ML algorithm looks for patterns in a data set with no pre-existing labels or human supervision. The algorithm learns about data points based on their relationship to other data points.
Can PCA and Cluster analysis be categorised as supervised or unsupervised learning?
Unsupervised as we don’t have labelled data, and try to say something about data points in their relationship to others.
What is the difference between unsupervised and supervised learning? (Range of problems and insights)
labaled and unlabeled data Unsupervised learning is open for a wider range of problems than supervised learning, but the insights we can gain are less powerful.
What are classification problems?
It is about labeling data, for example, a classifier that tells us if an image has a bird in it: it finds birst and non-bird images.
What is regression problems
Regression is about estimating continuous values, e.g given a set of features about a house, predict its price.
What is clustering problems about
Clustering is about a data point’s relation (e.g. distance) to other data points.
The bias-variance dilemma
bias-variance problem is the conflict in trying to simultaneously minimize the error in predictions on a training set, causes the model to have issues predicting outside the training set