Rationalizing Neural Predictions
Tao Lei, Regina Barzilay and Tommi Jaakkola Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology {taolei, regina,
Abstract
Prediction without justification has limited applicability. As a remedy, we learn to extract pieces of input text as justifications – rationales – that are tailored to be short and coherent, yet sufficient for making the same prediction. Our approach combines two modular components, generator and encoder, which are trained to operate well together. The generator specifies a distribution over text fragments as candidate rationales and these are passed through the encoder for prediction. Rationales are never given during training. Instead, the model is regularized by desiderata for rationales. We evaluate the approach on multi-aspect sentiment analysis against manually annotated test cases. Our approach outperforms attention-based baseline by a significant margin. We also successfully illustrate the method on the question retrieval task.111Our code and data are available at https://github.com/taolei87/rcnn.
中文速览
大白话来说，神经网络做预测时就像一个"黑箱"——你只知道它给出了结论，却不知道为什么，这在医疗等高风险场景里很成问题。为此，作者设计了一套"生成器＋编码器（generator + encoder）"双模块框架：生成器从原始文本中抠出一小段连续、简短的片段作为"理据（rationale）"，编码器再只看这段片段来做预测，两个模块联合训练、互相配合，全程不需要人工标注哪些词该被选中。在啤酒多维度情感评分和相似问题检索两项任务上测试，这套方法的理据抽取准确率达到96%，远超基于注意力机制（attention）的神经网络基线（81%）和传统SVM方法（38%）。这项工作的价值在于：它让神经模型在保持高预测性能的同时，能够用人类可读的文本片段解释自己的决策，朝着可解释人工智能迈出了切实一步。
原文 arXiv:1606.04155;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1606.04155v2