Incremental Permutation Feature Importance (iPFI): Towards Online Explanations on Data Streams
arXiv:2209.01939 · doi:10.1007/s10994-023-06385-y
Abstract
Explainable Artificial Intelligence (XAI) has mainly focused on static learning scenarios so far. We are interested in dynamic scenarios where data is sampled progressively, and learning is done in an incremental rather than a batch mode. We seek efficient incremental algorithms for computing feature importance (FI) measures, specifically, an incremental FI measure based on feature marginalization of absent features similar to permutation feature importance (PFI). We propose an efficient, model-agnostic algorithm called iPFI to estimate this measure incrementally and under dynamic modeling conditions including concept drift. We prove theoretical guarantees on the approximation quality in terms of expectation and variance. To validate our theoretical findings and the efficacy of our approaches compared to traditional batch PFI, we conduct multiple experimental studies on benchmark data with and without concept drift.
References in corpus (10)
- A Unified Approach to Interpreting Model Predictions
- All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously
- Correlation and variable importance in random forests
- Understanding Global Feature Contributions With Additive Importance Measures
- Feature relevance quantification in explainable AI: A causal problem
- Challenges in Benchmarking Stream Learning Algorithms with Real-world Data
- Grouped variable importance with random forests and application to multiple functional data analysis
- Visualizing the Feature Importance for Black Box Models
- Explaining by Removing: A Unified Framework for Model Explanation
- Model-agnostic Feature Importance and Effects with Dependent Features -- A Conditional Subgroup Approach