> The field of study that gives computers the ability to learn without being explicitly programmed > > <cite>Arthur Samuel (1959)</cite> > 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 $E$ > > <cite>Tom Mitchell (1997)</cite> ## Key Elements 1. Representation - How is the data represented? 2. Evaluation - What is the measure of performance? 3. Optimization - How can the performance be maximized? 4. Experimentation - Application of scientific method to multiple approaches ## [[Supervised Learning]] - [[Support Vector Machines (SVM)]] ## [[Unsupervised Learning]] - [[Gaussian Mixture Models]] - [[Expectation Maximization (EM) Algorithm]] - [[K-Means Clustering]] - [[Principal Component Analysis]] [[Backpropogation]]