liveBook · Manning

Pasting

As explained in Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI

Pasting is an ensemble technique that trains multiple models on subsamples drawn without replacement and combines their predictions.

Pasting is an averaging ensemble technique that trains multiple models on different subsamples of the training data, drawing examples without replacement. It pools regression predictions by averaging and classification predictions by majority vote, reducing variance while only partly increasing bias. section 5.1.1

Use pasting when comparing ensemble strategies for tabular prediction:

  • Prefer it when more reliable predictions matter and variance reduction is valuable. section 5.1.1
  • Distinguish it from bagging: pasting samples without replacement, whereas bagging bootstraps with replacement. section 5.6
  • Account for trade-offs: sampling can exclude important parts of the data distribution, increase bias, and make averaging methods computationally intensive. section 5.6

Pooling model predictions

For a regression task, average the predictions from models trained on separate subsamples. For a classification task, select the class receiving the majority of model votes. section 5.1.1

How Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI covers it

How deep it goes
Defines subsampling and prediction pooling, then contrasts bias, variance, computation, and data-distribution trade-offs.
Examples
Uses regression averaging and classification majority voting as the concrete prediction-pooling cases.
What it assumes
Assumes single-model learning and introduces ensemble averaging before comparing sampling strategies.
What you can do after
Evaluate whether without-replacement subsampling offers a useful variance reduction despite added bias and computation.
How it connects
Connects averaging with bagging, random subspaces, random patches, random forests, and boosting.
Cover of Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
liveBookmanning.com
Read Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI. The section links point to the parts of the book it draws on.