arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2203.04825cs.LGcs.CL
收藏

有效的子结构知识蒸馏

Efficient Sub-structured Knowledge Distillation

Tencent AI Lab(腾讯人工智能实验室);Tsinghua University(清华大学)

Wenye Lin, Yangming Li, Lemao Liu, Shuming Shi, Hai-tao Zheng

摘要:

结构化预测模型旨在解决一类问题,其中输出是复杂的结构,而不是单个变量。对这些模型执行知识蒸馏并不简单,因为它们的输出空间呈指数级大。在这项工作中,我们提出了一种方法,这是更简单的制定和更有效的培训比现有的方法。具体来说,我们通过在所有子结构上局部匹配他们的预测,而不是整个输出空间,将知识从教师模型转移到学生模型。通过这种方式,我们避免了采用一些耗时的技术,如动态规划(DP)解码输出结构,这允许并行计算,并使训练过程在实践中甚至更快。此外,它鼓励学生模型更好地模仿教师模型的内部行为。两个结构化预测任务的实验表明,我们的方法优于以前的方法和一半的时间成本为一个训练时期。

英文摘要:

Structured prediction models aim at solving a type of problem where the output is a complex structure, rather than a single variable. Performing knowledge distillation for such models is not trivial due to their exponentially large output space. In this work, we propose an approach that is much simpler in its formulation and far more efficient for training than existing approaches. Specifically, we transfer the knowledge from a teacher model to its student model by locally matching their predictions on all sub-structures, instead of the whole output space. In this manner, we avoid adopting some time-consuming techniques like dynamic programming (DP) for decoding output structures, which permits parallel computation and makes the training process even faster in practice. Besides, it encourages the student model to better mimic the internal behavior of the teacher model. Experiments on two structured prediction tasks demonstrate that our approach outperforms previous methods and halves the time cost for one training epoch.

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。

↑