Validation selection

As explained in Evolutionary Deep Learning: Genetic algorithms and neural networks

The process of selecting and applying procedures for evaluating model performance on training and testing data.

Validation selection chooses how to evaluate model performance, including training/testing data allocation and output visualization. Automating it reduces manual tuning and supports more explainable performance assessment as data changes. section 1.4.1

A validation-selection workflow can automate:

• Decisions about how much data to use for training and testing. section 1.4.1
• Visualization of a model’s output performance. section 1.4.1
• Recharacterization of model performance when the data changes. section 1.4.1

Figure 1.3 Steps for developing a good AI/ML model with AutoML and/or EDL

Use the AutoML/EDL development workflow to place validation selection alongside other model-building steps. section 1.4.1

How Evolutionary Deep Learning: Genetic algorithms and neural networks covers it

How deep it goes

Enumerates training/testing allocation and performance visualization but stops short of specifying particular validation algorithms or procedures.

Examples

Uses training/testing allocation, output-performance visualization, changing-data reassessment, and an AutoML/EDL workflow figure.

What you can do after

Apply it through AutoML to automate performance evaluation and support longer-term model explainability.

How it connects

Connects model performance with training/testing data, visualization, AutoML, hyperparameter optimization, model selection, and architecture.

liveBookmanning.com

Read Evolutionary Deep Learning: Genetic algorithms and neural networks on liveBook Get the book at manning.com

Related concepts

• AI/ML model development workflowpart of · An AI/ML model development workflow is the lifecycle of developing, deploying, monitoring, evaluating, and iterating an artificial-intelligence or machine-learning model.

This explainer was generated by Manning's knowledge graph from the text of Evolutionary Deep Learning: Genetic algorithms and neural networks. The section links point to the parts of the book it draws on.