AutoML

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

Automated machine learning is the automation of selecting, configuring, training, evaluating, and optimizing machine-learning models and workflows.

AutoML is a collection of tools, methods, and strategies for automating and enhancing AI/ML model development rather than a single technology. It formalizes model building so practitioners can move toward robust, commercializable systems, while evolutionary optimization provides one possible approach. section 1.4 section 1.4.1

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

Use the workflow diagram to navigate the stages AutoML can help automate across AI/ML model development. section 1.4.1

AutoML strategies can be applied progressively across the workflow:

• Optimize hyperparameters, data and features, model options, and model architecture. section 1.5.6
• Extend optimization to generative deep learning and deep reinforcement learning problems. section 1.5.6
• Combine evolutionary computation with deep learning through techniques such as neuroevolution and evolutionary hyperparameter optimization. section 1.5 section 1.6

AutoML platforms

Examples include DataRobot, Google Cloud AutoML, Amazon SageMaker AutoPilot, H2O AutoML, Azure Machine Learning, AutoKeras, and AutoTorch. These platforms automate different combinations of model development, tuning, data handling, or architecture search. section 1.4.1

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

How deep it goes

Defines AutoML and surveys workflow applications, but defers detailed implementation of its techniques to later treatment.

Examples

Uses commercial platforms and architecture-search tools including DataRobot, Google Cloud AutoML, AutoKeras, and AutoTorch.

What it assumes

Assumes familiarity with deep learning while teaching evolutionary computation basics from the ground up.

What you can do after

Apply evolutionary computation and deep learning together to optimize networks and select suitable techniques for problem classes.

How it connects

Connects AutoML with evolutionary computation, deep learning, hyperparameter optimization, feature engineering, model architecture, and reinforcement learning.

liveBookmanning.com

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

Related concepts

• hyperparameteruses · A hyperparameter is a setting chosen outside ordinary parameter learning that controls a model, algorithm, or learning procedure.
• Neural Architecture Searchalternative to · Neural architecture search is the automated exploration of neural-network architectures to find one that optimizes a specified objective or performs well on a task.
• feature engineeringuses · Feature engineering is the creation, transformation, or selection of input features from data to provide useful representations for analysis or modeling.
• machine learninguses · Machine learning is a field of computational methods in which systems learn patterns or relationships from data to make predictions, decisions, or other outputs.
• deep learningconfigures · Deep learning is a machine-learning approach that uses neural networks with multiple computational layers to learn representations from data.
• evolutionary computationincludes · Evolutionary computation is a family of search and optimization methods that iteratively improve candidate solutions through variation and selection inspired by biological evolution.

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.