{"id":16872,"date":"2016-12-26T23:23:30","date_gmt":"2016-12-26T21:23:30","guid":{"rendered":"http:\/\/hgpu.org\/?p=16872"},"modified":"2016-12-26T23:23:30","modified_gmt":"2016-12-26T21:23:30","slug":"language-modeling-with-gated-convolutional-networks","status":"publish","type":"post","link":"https:\/\/hgpu.org\/?p=16872","title":{"rendered":"Language Modeling with Gated Convolutional Networks"},"content":{"rendered":"<p>The pre-dominant approach to language modeling to date is based on recurrent neural networks. In this paper we present a convolutional approach to language modeling. We introduce a novel gating mechanism that eases gradient propagation and which performs better than the LSTM-style gating of (Oord et al, 2016) despite being simpler. We achieve a new state of the art on WikiText-103 as well as a new best single-GPU result on the Google Billion Word benchmark. In settings where latency is important, our model achieves an order of magnitude speed-up compared to a recurrent baseline since computation can be parallelized over time. To our knowledge, this is the first time a non-recurrent approach outperforms strong recurrent models on these tasks.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>The pre-dominant approach to language modeling to date is based on recurrent neural networks. In this paper we present a convolutional approach to language modeling. We introduce a novel gating mechanism that eases gradient propagation and which performs better than the LSTM-style gating of (Oord et al, 2016) despite being simpler. We achieve a new [&hellip;]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[11,89,3],"tags":[1782,14,1673,1815,20,67,1871,1881],"class_list":["post-16872","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-nvidia-cuda","category-paper","tag-computer-science","tag-cuda","tag-deep-learning","tag-nlp","tag-nvidia","tag-performance","tag-tesla-m40","tag-torch"],"views":6598,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/16872","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/users\/351"}],"replies":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=16872"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/16872\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=16872"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=16872"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=16872"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}