Editable Concept Bottleneck Models
KAUST · Shanghai Jiaotong University · The Hong Kong University of Science and Technology(Guang Zhou); The Hong Kong University of Science and Technology · HKUST(GZ) · Arizona State University · Stony Brook University · University of Auckland/RIKEN AIP · Hong Kong University of Science and Technology (Guangzhou)
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
Concept Bottleneck Models (CBMs) have garnered much attention for their ability to elucidate the prediction process through a human-understandable concept layer. However, most previous studies focused on cases where the data, including concepts, are clean. In many scenarios, we always need to remove/insert some training data or new concepts from trained CBMs due to different reasons, such as privacy concerns, data mislabelling, spurious concepts, and concept annotation errors. Thus, the challenge of deriving efficient editable CBMs without retraining from scratch persists, particularly in large-scale applications. To address these challenges, we propose Editable Concept Bottleneck Models (ECBMs). Specifically, ECBMs support three different levels of data removal: concept-label-level, concept-level, and data-level. ECBMs enjoy mathematically rigorous closed-form approximations derived from influence functions that obviate the need for re-training. Experimental results demonstrate the efficiency and effectiveness of our ECBMs, affirming their adaptability within the realm of CBMs.