google-deepmind/learning-to-learn
DeepMind's TensorFlow implementation of meta-learning where neural networks learn to optimize other neural networks.
Not currently ranked — collecting fresh signals.
star history
This repository implements the Learning to Learn meta-learning approach from the paper by Andrychowicz et al. It trains a learned optimizer (implemented as a neural network) that can learn to optimize other neural networks on various tasks. The framework includes support for simple quadratic functions, MNIST, and CIFAR-10 classification problems, allowing users to train and evaluate learned optimizers on standard benchmarks.
Frequently asked
- What is google-deepmind/learning-to-learn?
- DeepMind's TensorFlow implementation of meta-learning where neural networks learn to optimize other neural networks.
- Is learning-to-learn open source?
- Yes — google-deepmind/learning-to-learn is open source, released under the Apache-2.0 license.
- What language is learning-to-learn written in?
- google-deepmind/learning-to-learn is primarily written in Python.
- How popular is learning-to-learn?
- google-deepmind/learning-to-learn has 4.1k stars on GitHub.
- Where can I find learning-to-learn?
- google-deepmind/learning-to-learn is on GitHub at https://github.com/google-deepmind/learning-to-learn.