Most of the ML model techniques can be classified into three main types, i.e.
- The supervised learning model is trained on labelled data. It requires input-output pairs that include input data with corresponding labelled output data. Linear Regression, Neural Networks, SVM, Decision Trees, etc., are common supervised learning algorithms.
- An unsupervised learning model is trained on unlabeled data, where the algorithm tries to search out structures or patterns in the input data without explicit output labels. Principal Component Analysis, K-Means Clustering, Hierarchical Clustering, etc., are common unsupervised learning algorithms.
- The reinforcement Learning Model maximises a cumulative reward by including an agent learning to make decisions and act in an environment based on the feedback. Deep Q Networks (DQNs), Q-Learning, etc., are well-known reinforcement learning algorithms.
Other than the above mentioned, Transfer Learning Models, Generative Models, Ensemble Learning Models, and Instance-based Learning Models are some less popular machine learning models.