In Machine learning, ensembling is taking multiple models and combining their predictions into a single prediction with better accuracy.
There are different ways of doing this.
The simplest one is averaging. This can be used for regression tasks.
Let’s say you have two models. The prediction of the first model is P_1 and the prediction of the second model is P_2. Then the prediction of their ensemble can be \frac{P_1 + P_2}{2}.
This one is primarily useful for classification tasks. Each model in your ensemble predicts a class and your ensemble picks the majority choice.