Key-Value Memory Networks for Directly Reading Documents

Directly reading documents and being able to answer questions from them is a
key problem. To avoid its inherent difficulty, question answering (QA) has been
directed towards using Knowledge Bases (KBs) instead, which has proven
effective. Unfortunately KBs suffer from often being too restrictive, as the
schema cannot support certain types of answers, and too sparse, e.g. Wikipedia
contains much more information than Freebase. In this work we introduce a new
method, Key-Value Memory Networks, that makes reading documents more viable by
utilizing different encodings in the addressing and output stages of the memory
read operation. To compare using KBs, information extraction or Wikipedia
documents directly in a single framework we construct an analysis tool,
MovieQA, a QA dataset in the domain of movies. Our method closes the gap
between all three settings. It also achieves state-of-the-art results on the
existing WikiQA benchmark.

Source: http://bppro.link/?c=53N

No Comments

Leave a Reply

Training
The Savvy Cook – my second cookbook!

(UK cover on the left, US cover on the right) I’m done with exams, back in London and I’m finally going to talk about my new book, The Savvy Cook, today!! If you haven’t pre-ordered already you can do so here: Amazon UK   (out June 29th) Amazon US   (out June 20th) Dymocks …

Training
Young Gun-founded Entrepreneur First clinches £9.5m from LinkedIn co-founder

Latest funding will be used to help the organisation scale and to invest in its ongoing operational costs…In Funding news Source: http://bppro.link/?c=VRx

Training
Quite intimidating. Much. Very.

Share and Enjoy:The post Quite intimidating. Much. Very. appeared first on Indexed. Source: http://bppro.link/?c=TDF