arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1810.09309cs.CLcs.LG
收藏

实时的神经输入法

Real-time Neural-based Input Method

Carnegie Mellon University(卡内基梅隆大学);Microsoft(微软);The University of Tokyo(东京大学)

Jiali Yao, Raphael Shu, Xinjian Li, Katsutoshi Ohtsuki, Hideki Nakayama

摘要:

输入法是每个移动和桌面设备上提供文本建议的基本服务。它将连续的键盘输入转换为目标语言中的字符,这是日语和汉语用户不可缺少的。由于关键的资源限制和有限的网络带宽的目标设备,应用神经模型的输入法没有得到很好的探索。在这项工作中,我们应用基于LSTM的语言模型的输入法,并评估其性能的预测和转换任务与日本BCCWJ语料库。我们明确指出,瓶颈是转换过程中缓慢的softmax计算。为了解决这个问题,我们提出了增量softmax近似方法,该方法使用选定的子集词汇计算softmax,并在未来步骤中更新词汇时固定陈旧概率。我们将这种方法称为增量选择性softmax。实验结果表明,当转换大词汇量的日语输入序列时,softmax算法的计算速度提高了2个数量级,在普通CPU上达到了实时的【形】速度。我们还利用模型压缩的潜力,实现了92%的模型大小减少,而不损失准确性。

英文摘要:

The input method is an essential service on every mobile and desktop devices that provides text suggestions. It converts sequential keyboard inputs to the characters in its target language, which is indispensable for Japanese and Chinese users. Due to critical resource constraints and limited network bandwidth of the target devices, applying neural models to input method is not well explored. In this work, we apply a LSTM-based language model to input method and evaluate its performance for both prediction and conversion tasks with Japanese BCCWJ corpus. We articulate the bottleneck to be the slow softmax computation during conversion. To solve the issue, we propose incremental softmax approximation approach, which computes softmax with a selected subset vocabulary and fix the stale probabilities when the vocabulary is updated in future steps. We refer to this method as incremental selective softmax. The results show a two order speedup for the softmax computation when converting Japanese input sequences with a large vocabulary, reaching real-time speed on commodity CPU. We also exploit the model compressing potential to achieve a 92% model size reduction without losing accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑