GAM Changer: Editing Generalized Additive Models with Interactive Visualization
This addresses the need for responsible ML deployment by allowing users to align model behaviors with their knowledge and values, though it is incremental as it builds on existing GAM frameworks.
The paper tackles the problem of fixing undesirable patterns in interpretable machine learning models by introducing GAM Changer, an interactive system that enables data scientists and domain experts to edit Generalized Additive Models through visualization, resulting in a tool that runs locally without extra compute resources.
Recent strides in interpretable machine learning (ML) research reveal that models exploit undesirable patterns in the data to make predictions, which potentially causes harms in deployment. However, it is unclear how we can fix these models. We present our ongoing work, GAM Changer, an open-source interactive system to help data scientists and domain experts easily and responsibly edit their Generalized Additive Models (GAMs). With novel visualization techniques, our tool puts interpretability into action -- empowering human users to analyze, validate, and align model behaviors with their knowledge and values. Built using modern web technologies, our tool runs locally in users' computational notebooks or web browsers without requiring extra compute resources, lowering the barrier to creating more responsible ML models. GAM Changer is available at https://interpret.ml/gam-changer.