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 1506.03099cs.LGcs.CLcs.CV
收藏

基于循环神经网络的序列预测定时采样

Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks

Google Research(谷歌研究院)

Samy Bengio, Oriol Vinyals, Navdeep Jaitly, Noam Shazeer

摘要:

循环神经网络可以被训练来在给定一些输入的情况下产生令牌序列,如最近在机器翻译和图像描述中的结果所例示的。目前训练它们的方法包括在给定当前(循环)状态和先前令牌的情况下,最大化序列中每个令牌的可能性。在推理时,未知的先前标记然后被模型本身生成的标记替换。训练和推理之间的这种差异可能会产生错误,这些错误可能会沿着生成的序列快速积累。我们提出了一个课程学习策略,轻轻地改变训练过程,从一个完全指导计划使用真正的以前的令牌,对一个较少的指导计划,主要是使用生成的令牌。在几个序列预测任务上的实验表明,该方法产生了显着的改善。此外,它还成功用于我们2015年MSCOCO图像描述挑战赛的获奖作品。

英文摘要:

Recurrent Neural Networks can be trained to produce sequences of tokens given some input, as exemplified by recent results in machine translation and image captioning. The current approach to training them consists of maximizing the likelihood of each token in the sequence given the current (recurrent) state and the previous token. At inference, the unknown previous token is then replaced by a token generated by the model itself. This discrepancy between training and inference can yield errors that can accumulate quickly along the generated sequence. We propose a curriculum learning strategy to gently change the training process from a fully guided scheme using the true previous token, towards a less guided scheme which mostly uses the generated token instead. Experiments on several sequence prediction tasks show that this approach yields significant improvements. Moreover, it was used successfully in our winning entry to the MSCOCO image captioning challenge, 2015.

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。

↑