{"author_url":"https://blog.hatena.ne.jp/Ryobot/","provider_name":"Hatena Blog","blog_title":"\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u30d6\u30ed\u30b0","version":"1.0","image_url":"https://cdn-ak.f.st-hatena.com/images/fotolife/R/Ryobot/20171221/20171221145207.png","provider_url":"https://hatena.blog","description":"ConvS2S \u306f\u8131 RNN \u306b\u3088\u308b\u7cfb\u5217\u65b9\u5411\u306e\u4e26\u5217\u5316\u3068\u3044\u3046\u98a8\u6f6e\u3092\u6c7a\u5b9a\u3065\u3051\u305f\u30cb\u30e5\u30fc\u30e9\u30eb\u6a5f\u68b0\u7ffb\u8a33\u306e\u624b\u6cd5\u3067\u3059\uff0e\u5358\u8a9e\u5217\u306e\u51e6\u7406\u3092 LSTM \u304b\u3089 CNN \u306b\u7f6e\u304d\u63db\u3048\uff0cGNMT \u3088\u308a\u9ad8\u6027\u80fd\u304b\u3064\u8a13\u7df4\u3082 5 \u500d\u9ad8\u901f\u3092\u9054\u6210\u3057\u3066\u3044\u307e\u3059\uff0eLSTM \u306f\u30b2\u30fc\u30c6\u30a3\u30f3\u30b0\u304c\u91cd\u8981\u306a\u306e\u3067\uff0c\u3053\u306e\u90e8\u5206\u3092\u629c\u304d\u51fa\u3057\u305f Gated Linear Unit \u3092\u4f75\u7528\u3057\u307e\u3059\uff0e","type":"rich","categories":["Deep Learning","Neural Machine Translation"],"blog_url":"https://deeplearning.hatenablog.com/","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fdeeplearning.hatenablog.com%2Fentry%2Fconvs2s\" title=\"\u8ad6\u6587\u89e3\u8aac Convolutional Sequence to Sequence Learning (ConvS2S) - \u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u30d6\u30ed\u30b0\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","title":"\u8ad6\u6587\u89e3\u8aac Convolutional Sequence to Sequence Learning (ConvS2S)","published":"2017-12-20 18:40:03","url":"https://deeplearning.hatenablog.com/entry/convs2s","width":"100%","author_name":"Ryobot","height":"190"}