Meta-learning in Reinforcement Learning

By Nicolas Schweighofer, Kenji Doya. 2003

275 citations recorded by the source corpus.

Subfields: Cellular and Molecular Neuroscience, Cognitive Neuroscience, Reinforcement Learning, Meta & Continual Learning

How researchers see this paper

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

In Cellular and Molecular Neuroscience

In Cognitive Neuroscience

In Reinforcement Learning

In Meta & Continual Learning

Citations in the corpus

Learning to reinforcement learn (2016, background): “…for meeting these challenges, which we refer to as deep meta-reinforcement learning , a label that is intended to both link it with and distinguish it from previous work employing the term “meta-reinforcement learning” (e.g. Schmidhuber et al., 1996; Schweighofer and Doya, 2003, discussed later).”
One-shot Learning with Memory-Augmented Neural Networks (2016, methodology): “Although the term has been used in numerous senses (Schmidhuber et al., 1997; Caruana, 1997; Schweighofer & Doya, 2003; Brazdil et al., 2003), meta-learning generally refers to a scenario in which an agent learns at two levels, each associated with different time scales.”
Meta-Learning in Neural Networks: A Survey (2020, neutral): “lowed by various extensions [54], [55].”
A neural circuit model of flexible sensorimotor mapping: learning and forgetting on multiple timescales. (2007, background): “Memory performances are higher with modifiable learning rates than with a static distribution of learning rates because each synapse changes adaptively the rules by which it is modified (metalearning) (Schweighofer and Doya, 2003; Soltani et al., 2006).”
Online meta-learning by parallel algorithm competition (2017, methodology): “ry approach (i.e., the learning started from scratch in each generation) to ﬁnd appropriate ﬁxed values of the meta-parameters (Unemi et al., 1994; Eriksson et al., 2003; Elfwing et al., 2008, 2011). Schweighofer and Doya (2003) proposed a meta-learning method based on the mid-term and the long-term”
Meta-Learning with Memory-Augmented Neural Networks (2016, methodology): “Although the term has been used in numerous senses (Schmidhuber et al., 1997; Caruana, 1997; Schweighofer & Doya, 2003; Brazdil et al., 2003), meta-learning generally refers to a scenario in which an agent learns at two levels, each associated with different time scales.”
Neural mechanism for stochastic behaviour during a competitive game (2006, methodology): “In reinforcement learning, meta-learning algorithms are commonly evoked to adjust parameters, such as the learning rate for updating value functions and the rate of temporal discounting for delayed rewards (Doya, 2002; Schweighofer & Doya, 2003).”
A Meta-learning Method Based on Temporal Difference Error (2009, methodology): “proposed the meta-learning method based on mid-term and longterm rewards [4].”
Reinforcement Learning, Fast and Slow. (2019, unknown): “Meta-learning occurs when one learning system progressively adjusts the operation of a second learning system, such that the latter operates with increasing speed and efficiency [83, 84, 85, 86, 87].”