Work

Towards learning to explain with concept bottleneck models: mitigating information leakage

Year: 2022

Type: preprint

Abstract: Concept bottleneck models perform classification by first predicting which of a list of human provided concepts are true about a datapoint. Then a downstream model uses these predicted concept labels to predict the target label. The predicted concepts act as a rationale for the target prediction. Model trust issues emerge in this paradigm when soft concept labels are used: it has previously been observed that extra information about the data distribution leaks into the concept predictions. In this work we show how Monte-Carlo Dropout can be used to attain soft concept predictions that do not contain leaked information.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2211.03656

Authors: Lockhart, Joshua, Marchesotti, Nicolas, Magazzeni, Daniele, Veloso, Manuela

Language: English

Topic: Explainable Artificial Intelligence (XAI)

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W4308618482

Cites:

Cited by:

Related to: 10