Audio player

117 - Interpreting NLP Model Predictions, with Sameer Singh

Description

We interviewed Sameer Singh for this episode, and discussed an overview of recent work in interpreting NLP model predictions, particularly instance-level interpretations. We started out by talking about why it is important to interpret model outputs and why it is a hard problem. We then dove into the details of three kinds of interpretation techniques: attribution based methods, interpretation using influence functions, and generating explanations. Towards the end, we spent some time discussing how explanations of model behavior can be evaluated, and some limitations and potential concerns in evaluation methods. Sameer Singh is an Assistant Professor of Computer Science at the University of California, Irvine. Some of the techniques discussed in this episode have been implemented in the AllenNLP Interpret framework (details and demo here: https://allennlp.org/interpret).

Comments

Listen to 120 - Evaluation of Text Generation, with Asli Celikyilmaz

120 - Evaluation of Text Generation, with Asli Celikyilmaz

2020-10-02
0
Listen to 119 - Social NLP, with Diyi Yang

119 - Social NLP, with Diyi Yang

2020-09-03
0
Listen to 118 - Coreference Resolution, with Marta Recasens

118 - Coreference Resolution, with Marta Recasens

2020-08-26
0
Listen to 116 - Grounded Language Understanding, with Yonatan Bisk

116 - Grounded Language Understanding, with Yonatan Bisk

2020-07-02
0
Listen to 115 - AllenNLP, interviewing Matt Gardner

115 - AllenNLP, interviewing Matt Gardner

2020-06-17
0