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

A fundamental task for artificial intelligence is learning. Deep NeuralNetworks have proven to cope perfectly with all learning paradigms, i.e.supervised, unsupervised, and reinforcement learning. Nevertheless, traditionaldeep learning approaches make use of cloud computing facilities and do notscale well to autonomous agents with low computational resources. Even in thecloud, they suffer from computational and memory limitations, and they cannotbe used to model adequately large physical worlds for agents which assumenetworks with billions of neurons. These issues are addressed in the last fewyears by the emerging topic of sparse training, which trains sparse networksfrom scratch. This paper discusses sparse training state-of-the-art, itschallenges and limitations while introducing a couple of new theoreticalresearch directions which has the potential of alleviating sparse traininglimitations to push deep learning scalability well beyond its currentboundaries. Nevertheless, the theoretical advancements impact in complexmulti-agents settings is discussed from a real-world perspective, using thesmart grid case study.

Sparse Training Theory for Scalable and Efficient Agents