Fine-grained Interpretation and Causation Analysis in Deep NLP Models
It addresses the problem of model interpretability for researchers and practitioners in NLP, but is incremental as it summarizes existing work in a tutorial format.
This tutorial paper presents methods for interpreting fine-grained components of deep NLP models, focusing on analyzing individual neurons and input features to explain model decisions, and introduces toolkits like NeuroX and Captum to support these functionalities.
This paper is a write-up for the tutorial on "Fine-grained Interpretation and Causation Analysis in Deep NLP Models" that we are presenting at NAACL 2021. We present and discuss the research work on interpreting fine-grained components of a model from two perspectives, i) fine-grained interpretation, ii) causation analysis. The former introduces methods to analyze individual neurons and a group of neurons with respect to a language property or a task. The latter studies the role of neurons and input features in explaining decisions made by the model. We also discuss application of neuron analysis such as network manipulation and domain adaptation. Moreover, we present two toolkits namely NeuroX and Captum, that support functionalities discussed in this tutorial.