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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Can neural nets learn logic? We approach this classic question with currentmethods, and demonstrate that recurrent neural networks can learn to recognizefirst order logical entailment relations between expressions. We define anartificial language in first-order predicate logic, generate a large dataset ofsample 'sentences', and use an automatic theorem prover to infer the relationbetween random pairs of such sentences. We describe a Siamese neuralarchitecture trained to predict the logical relation, and experiment withrecurrent and recursive networks. Siamese Recurrent Networks are surprisinglysuccessful at the entailment recognition task, reaching near perfectperformance on novel sentences (consisting of known words), and evenoutperforming recursive networks. We report a series of experiments to test theability of the models to perform compositional generalization. In particular,we study how they deal with sentences of unseen length, and sentencescontaining unseen words. We show that set-ups using LSTMs and GRUs obtain highscores on these tests, demonstrating a form of compositionality.

Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization