Vertex AI AutoML
AutoML tools in Google Cloud have existed a lot longer than Vertex AI, which was launched to primarily unify most of the separate ML offerings existing in GCP. GCP AutoML makes use of models such as NASNet and constantly benefits from the AI research happening in other divisions of the Alphabet teams, such as Google DeepMind. Few of the interesting papers on the topic are listed below:
- Learning Transferable Architectures for Scalable Image Recognition. https://arxiv.org/pdf/1707.07012.pdf
- Regularized Evolution for Image Classifier Architecture Search. https://arxiv.org/pdf/1802.01548.pdf
- Large-Scale Evolution of Image Classifiers. https://arxiv.org/pdf/1703.01041.pdf
The use cases supported by Vertex AI AutoML are as follows:
- Tabular data:
- Classification (an example is covered in this chapter)
- Regression
- Time series forecasting
- Image data:
- Image classification (an example is covered in Chapter 8)
- Object detection
- Natural language:
- Text...
