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

In order for artificial neural networks to begin accurately mimickingbiological ones, they must be able to adapt to new exigencies withoutforgetting what they have learned from previous training. Lifelong learningapproaches to artificial neural networks attempt to strive towards this goal,yet have not progressed far enough to be realistically deployed for naturallanguage processing tasks. The proverbial roadblock of catastrophic forgettingstill gate-keeps researchers from an adequate lifelong learning model. Whileefforts are being made to quell catastrophic forgetting, there is a lack ofresearch that looks into the importance of class ordering when training on newclasses for incremental learning. This is surprising as the ordering of"classes" that humans learn is heavily monitored and incredibly important.While heuristics to develop an ideal class order have been researched, thispaper examines class ordering as it relates to priming as a scheme forincremental class learning. By examining the connections between variousmethods of priming found in humans and how those are mimicked yet remainunexplained in life-long machine learning, this paper provides a betterunderstanding of the similarities between our biological systems and thesynthetic systems while simultaneously improving current practices to combatcatastrophic forgetting. Through the merging of psychological priming practiceswith class ordering, this paper is able to identify a generalizable method forclass ordering in NLP incremental learning tasks that consistently outperformsrandom class ordering.

Utilizing Priming to Identify Optimal Class Ordering to Alleviate Catastrophic Forgetting