Metalearning

By Tom Schaul, Jürgen Schmidhuber. 2010

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep learning in neural networks: An overview (2014, methodology): “8 Universal RL General purpose learning algorithms may improve themselves in open-ended fashion and environment-specific ways in a lifelong learning context (Schmidhuber, 1987; Schmidhuber et al., 1997b,a; Schaul and Schmidhuber, 2010).”
On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models (2015, background): “, neural pushdown automata [47, 175], NNs with quickly modifiable, differentiable external memory based on fast weights [229], or closely related RNN-based meta-learners [232, 231, 103, 219].”
Few-Shot Learning via Learning the Representation, Provably (2020, background): “Representation learning is also closely connected to meta-learning (Schaul and Schmidhuber, 2010).”
One Big Net For Everything (2018, methodology): “Many RNN-like models can be used to build general computers, e.g., RNNs controlling push-down automata [9, 41] or other types of differentiable memory [20] including differentiable fast weights [63, 65], as well as closely related RNN-based meta-learners [66, 25, 51].”
Meta Reinforcement Learning with Latent Variable Gaussian Processes (2018, methodology): “Meta learning has long been proposed as a form of learning that would allow systems to systematically build up and re-use knowledge across different but related tasks [26, 32].”
Unicorn: Continual Learning with a Universal, Off-policy Agent (2018, background): “Closely related to the choice of how to act in the presence of many goals is the choice of what (pseudo-rewards) to learn about, which can be seen as a form of learning to learn, or meta-learning [38].”
Learning Transferable Architectures for Scalable Image Recognition (2017, background): “The concept of having one neural network interact with a second neural network to aid the learning process, or learning to learn or meta-learning [23, 48] has attracted much attention in recent years [1, 61, 14, 19, 34, 44, 15].”
Machine Learning - The Art and Science of Algorithms that Make Sense of Data (2012, background): “Bagging trains diverse models from samples of the training data, and was introduced by Breiman (1996a). Random forests, usually attributed to Breiman (2001), combine bagged decision trees with random subspaces; similar ideas were developed by Ho (1995) and Amit and Geman (1997).”
A Survey of Optimization Methods From a Machine Learning Perspective (2019, background): “Meta learning [45], [46] has recently become very popular in the field of machine learning.”