What are the learning methods of machine learning?

1. supervised learning

Under supervised learning, the input data is called "training data", and each group of training data has clear identification or results, such as "garbage" and "non-garbage" in the anti-garbage system and "1", "2", "3" and "4" in handwritten numeral recognition.

When establishing a prediction model, supervised learning establishes a learning process, compares the prediction results with the actual results of "training data", and constantly adjusts the prediction model until the prediction results of the model reach an expected accuracy.

Classification problems, regression problems and other common application scenarios of supervised learning. Common algorithms include logistic regression and back propagation neural network.

Strengthen study

In this learning mode, the input data is used as feedback to the model, instead of just being used as a way to check whether the model is right or wrong, like the supervised model. Under reinforcement learning, the input data is directly fed back to the model, and the model must be adjusted immediately.

Common application scenarios include dynamic systems and robot control. Common algorithms are Q learning and time difference learning.

3. Unsupervised learning In unsupervised learning, the data is not specially marked, and the learning model is to infer some internal structures of the data. Common application scenarios include learning association rules and clustering. The common algorithms are Apriori algorithm and k-Means algorithm.

4. Semi-supervised learning

In this learning mode, some input data are recognized and some are not. This learning model can be used for forecasting, but the model needs to learn the internal structure of data first in order to organize the data reasonably for forecasting.

Application scenarios include classification and regression, and algorithms include some extensions of commonly used supervised learning algorithms. These algorithms first try to model the unlabeled data, and then predict the tagged data. Such as graphic reasoning algorithm or Laplacian SVM.