# Ablative analysis

**ablative analysis**

Ablative analysis is a technique for evaluating machine learning models, to help an analyst make tradeoffs between accuracy and complexity.

Suppose a given model, or hypothesis, , shows acceptably low error after being trained on empirical inputs and the associated output ground truth. Here, is an matrix of input examples, each having scalar features, and is a column vector of scalar outputs. If the model is, e.g., a binary classifier, then each output will be 0 or 1.

We then input some brand new data and record 's generalization error, or the RMS difference between and the empirically measured ground truth.

There are several motivations for simplifying while achieving similar error: we may be able to reduce the cost of gathering input data by using fewer input features, and we may be able to use a machine learning algorithm with reduced complexity for training or for predictions, in terms of space or time, perhaps by shrinking a processing pipeline buried within . Also, a simpler model may avoid over-fitting and exhibit better generalization error.

Ablative analysis is a technique for generating such models, which will compete with . Choose a column of the feature vector to discard, and use the modified to train a new model . Now record 's generalization error, the RMS difference between its output and . If the error is still acceptably low we might choose to simplify our model, adopting the model that does not depend on the uninformative feature we discarded.

Alternatively, suppose that all features are necessary for acceptable performance, and that contains an ensemble of models, or a pipeline of several processing stages. A stage might clip values to a range, or condition one variable upon another. As a model is developed and debugged, such special cases may accrete over time. Choose an ensemble member or a stage, and discard it, to form model . As before if its measured error is acceptably low we might prefer it over . For the features or stages we choose to retain, we can now quantify how much they contribute to the success of , relative to other parts of the model.

## References[edit]

Hastie: 7.2 Bias, Variance and Model Complexity

## External links[edit]

- Andrew Ng, Advice for applying Machine Learning [1], pp. 23-24.

- Held, Thrun, and Savarese, Learning to Track at 100 FPS with Deep Regression Networks [2] section 6.4.

- Chen Sun et al., Revisiting Unreasonable Effectiveness of Data in Deep Learning Era [3]

- Minghuang Ma et al., Going Deeper into First-Person Activity Recognition [4]

- Sara Maatta, Predicting groundwater levels using linear regression and neural networks [5]

- Jan Overgoor et al., Predicting Negative CouchSurfing Experiences using Local Information [6]

- Pamela Bhattacharya et al., Automated, highly-accurate, bug assignment using machine learning and tossing graphs [7]

## ablative analysis[edit]

This article "Ablative analysis" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:Ablative analysis. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.