Feb. 12, 2024, 5:41 a.m. | Ao Sun Yuanyuan Yuan Pingchuan Ma Shuai Wang

cs.LG updates on arXiv.org arxiv.org

Concept Bottleneck Models (CBMs) aim to deliver interpretable and interventionable predictions by bridging features and labels with human-understandable concepts. While recent CBMs show promising potential, they suffer from information leakage, where unintended information beyond the concepts (either when concepts are represented with probabilities or binary states) are leaked to the subsequent label prediction. Consequently, distinct classes are falsely classified via indistinguishable concepts, undermining the interpretation and intervention of CBMs.
This paper alleviates the information leakage issue by introducing label supervision …

aim beyond binary concept concepts cs.ai cs.lg features hierarchical human information labels predictions show

Research Scholar (Technical Research)

@ Centre for the Governance of AI | Hybrid; Oxford, UK

HPC Engineer (x/f/m) - DACH

@ Meshcapade GmbH | Remote, Germany

Business Intelligence Analyst Lead

@ Zillow | Mexico City

Lead Data Engineer

@ Bristol Myers Squibb | Hyderabad

Big Data Solutions Architect

@ Databricks | Munich, Germany

Senior Data Scientist - Trendyol Seller

@ Trendyol | Istanbul (All)