Evolving Echo State Networks for Minimally Cognitive Unsupervised Learning Tasks
This thesis investigates the use of Echo State Networks (ESNs) in unsupervised learning environments, by
employing Evolutionary Algorithms (EAs) to evolve ESNs to control an agent that performs a novel,
minimally-cognitive learning task. The task employed in this thesis is a modified version of the classic
video game Frogger. ESNs are investigated since they promise to combine the temporal abilities seen in other
Recurrent Neural Networks (RNNs) with a straightforward method to train the network. However, previous
work employing ESNs in unsupervised environments is lacking.
The evolved ESNs are compared to feed-forward Artificial Neural Networks (ANNs) as well as ESNs trained
with regular supervised learning, in a comparative performance measure to find out which method is best
suited to control Frogger.
The results from this thesis show that not only do ESNs work well with EAs, but they surpass traditional
feed-forward ANNs on the Frogger task. Additionally, it is shown that for the Frogger task, evolved ESNs
also outperform ESNs trained with supervised learning. The results from this work should serve as an
encouragement to use ESNs for more tasks in the future, due to their competitive performance and ease of setup.
Place, publisher, year, edition, pages
Institutt for datateknikk og informasjonsvitenskap , 2013. , 100 p.
IdentifiersURN: urn:nbn:no:ntnu:diva-22966Local ID: ntnudaim:8904OAI: oai:DiVA.org:ntnu-22966DiVA: diva2:655600
Downing, Keith, Professor