English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Memory-augmented neural networks (MANNs) can solve algorithmic tasks likesorting. However, they often do not generalize to lengths of input sequencesnot seen in the training phase. Therefore, we introduce two approachesconstraining the state-space of the network controller to improve thegeneralization to out-of-distribution-sized input sequences: state compressionand state regularization. We show that both approaches can improve thegeneralization capability of a particular type of MANN, the differentiableneural computer (DNC), and compare our approaches to a stateful and a statelesscontroller on a set of algorithmic tasks. Furthermore, we show that especiallythe combination of both approaches can enable a pre-trained DNC to be extendedpost hoc with a larger memory. Thus, our introduced approaches allow to train aDNC using shorter input sequences and thus save computational resources.Moreover, we observed that the capability for generalization is oftenaccompanied by loop structures in the state-space, which could correspond tolooping constructs in algorithms.

State-Space Constraints Improve the Generalization of the Differentiable Neural Computer in some Algorithmic Tasks