Continual Learning through Evolvable Neural Turing Machines
- Benno Lüders,
- Mikkel Schläger,
Research Output:
Contribution to conference - NOT published in proceeding or journal
Paper
Peer-review
Open access
Publication Information
Output type
Research Output:
Contribution to conference - NOT published in proceeding or journal
Paper
Peer-reviewOriginal language
EnglishPublication milestones
- Published - 2016
Publication status
Published - 2016
Abstract
Continual learning, i.e. the ability to sequentially learn tasks without catastrophic
forgetting of previously learned ones, is an important open challenge in machine
learning. In this paper we take a step in this direction by showing that the recently
proposed Evolving Neural Turing Machine (ENTM) approach is able to perform
one-shot learning in a reinforcement learning task without catastrophic forgetting
of previously stored associations.
forgetting of previously learned ones, is an important open challenge in machine
learning. In this paper we take a step in this direction by showing that the recently
proposed Evolving Neural Turing Machine (ENTM) approach is able to perform
one-shot learning in a reinforcement learning task without catastrophic forgetting
of previously stored associations.
Access to documents
Accepted author manuscript
Accepted author manuscript, 526.04 KB