Meta Learning

In Machine Learning there are so-called meta-tasks:


Meta Learning is a set of Machine Learning techniques for addressing these tasks. The most popular are

These techniques generate samples from the data and then train and evaluate models based on these samples

They all have two common steps:

  • samples are generated from the input data
  • Machine Learning models are trained on these samples


Scalable Meta Learning

See the paper by S. Schelter:

  • Schelter, Sebastian, et al. "Efficient Sample Generation for Scalable Meta Learning." (pdf


Links


Sources

  • Schelter, Sebastian, et al. "Efficient Sample Generation for Scalable Meta Learning." (pdf poster)
Share your opinion