Evolving Neural Turing Machines for Reward-based Learning
By Rasmus Boll Greve, Emil Juul Jacobsen, Sebastian Risi. 2016
42 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Electrical and Electronic Engineering
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sebastian Risi
- Andrea Soltoggio
- Kenneth O. Stanley
- Benno Lüders
- Mikkel Schläger
- Jeff Clune
- Aleksandra Korach
- Joel Lehman
In Electrical and Electronic Engineering
- Sebastian Risi
- Andrea Soltoggio
- Kenneth O. Stanley
- Benno Lüders
- Mikkel Schläger
- Aleksandra Korach
- Jeff Clune
- Jean-Baptiste Mouret
Citations in the corpus
- Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks (2017, methodology): “EPANNs need not be differentiable, as in gradient descent based approaches, thereby allowing a greater variety in architectures (Greve et al., 2016).”
- Continual and One-Shot Learning Through Neural Networks with Dynamic External Memory (2017, methodology): “Moving beyond the original demonstration of the ENTM [21] in the T-Maze domain, which did not require the agent to learn new associations during its lifetime, this paper shows how the ENTM is able to overcome catastrophic forgetting in a reinforcement learning task.”
- Continual Learning through Evolvable Neural Turing Machines (2016, methodology): “The ENTM often finds compact network topologies to solve a particular task, thereby avoiding searching through unnecessarily high-dimensional spaces [6].”