Learning Dynamic Feature Selection for Fast Sequential Prediction
Learning Dynamic Feature Selection for Fast Sequential Prediction
复制标题
学习动态特征选择以实现快速序列预测
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
A. McCallum
中科院分区：
文献类型：
--
作者：
Emma Strubell;L. Vilnis;Kate Silverstein;A. McCallum
我们提出了配对学习和推理算法，用于显着减少计算和提高分类器中向量点积的速度，这些分类器是许多NLP组件的核心。这是通过将特征划分为一系列模板来实现的，这些模板是有序的，使得通常仅使用所有特征的一小部分就可以达到高置信度。安排参数估计以最大化该序列中的准确度和早期置信度。我们的方法比其他相关的级联方法更简单，更适合于NLP。我们目前的实验，从左到右的词性标注，命名实体识别和过渡为基础的依赖分析。在典型的基准测试数据集上，我们可以保持POS标记准确率高于97%，解析LAS高于88.5%，运行时间减少了五倍以上，NER F1高于88，速度提高了2倍以上。
We present paired learning and inference algorithms for significantly reducing computation and increasing speed of the vector dot products in the classifiers that are at the heart of many NLP components. This is accomplished by partitioning the features into a sequence of templates which are ordered such that high confidence can often be reached using only a small fraction of all features. Parameter estimation is arranged to maximize accuracy and early confidence in this sequence. Our approach is simpler and better suited to NLP than other related cascade methods. We present experiments in left-to-right part-of-speech tagging, named entity recognition, and transition-based dependency parsing. On the typical benchmarking datasets we can preserve POS tagging accuracy above 97% and parsing LAS above 88.5% both with over a five-fold reduction in run-time, and NER F1 above 88 with more than 2x increase in speed.