arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1902.11269cs.CLcs.LG
收藏

无Softmax层的高效上下文表示学习

Efficient Contextual Representation Learning Without Softmax Layer

Peking University(北京大学);University of California, Los Angeles(加州大学洛杉矶分校)

Liunian Harold Li, Patrick H. Chen, Cho-Jui Hsieh, Kai-Wei Chang

摘要:

上下文表示模型在改进各种下游任务方面取得了巨大成功。然而,这些基于语言模型的编码器由于大的参数大小和高的计算复杂性而难以训练。通过仔细检查训练过程,我们发现softmax层(输出层)由于词汇量大而导致效率低下。因此,我们重新设计了学习目标,并提出了一个有效的框架,训练上下文表示模型。具体而言,所提出的方法通过执行具有降维的语言建模来绕过softmax层,并允许模型利用预先训练的词嵌入。我们的框架将花费在输出层上的时间减少到可以忽略的水平,消除了softmax层的几乎所有可训练参数,并在不截断词汇的情况下进行语言建模。当应用于ELMo时,我们的方法实现了4倍的加速比,并消除了80%的可训练参数,同时在下游任务上实现了有竞争力的性能。

英文摘要:

Contextual representation models have achieved great success in improving various downstream tasks. However, these language-model-based encoders are difficult to train due to the large parameter sizes and high computational complexity. By carefully examining the training procedure, we find that the softmax layer (the output layer) causes significant inefficiency due to the large vocabulary size. Therefore, we redesign the learning objective and propose an efficient framework for training contextual representation models. Specifically, the proposed approach bypasses the softmax layer by performing language modeling with dimension reduction, and allows the models to leverage pre-trained word embeddings. Our framework reduces the time spent on the output layer to a negligible level, eliminates almost all the trainable parameters of the softmax layer and performs language modeling without truncating the vocabulary. When applied to ELMo, our method achieves a 4 times speedup and eliminates 80% trainable parameters while achieving competitive performance on downstream tasks.

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。

↑