Learning to Intervene on Concept Bottlenecks

arXiv:2308.13453v2 Announce Type: replace
Abstract: While traditional deep learning models often lack interpretability, concept bottleneck models (CBMs) provide inherent explanations via their concept representations. Specifically, they allow users to perform interventional interactions on these concepts by updating the concept values and thus correcting the predictive output of the model. Traditionally, however, these interventions are applied to the model only once and discarded afterward. To rectify this, we present concept bottleneck memory models (CB2M), an extension to CBMs. Specifically, a CB2M …

abstract arxiv bottlenecks concept concepts cs.ai cs.lg deep learning however interactions interpretability predictive type values via