DeepAI AI Chat
Log In Sign Up

Incremental Permutation Feature Importance (iPFI): Towards Online Explanations on Data Streams

by   Fabian Fumagalli, et al.
Bielefeld University
Universität München

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.


page 17

page 18


iSAGE: An Incremental Version of SAGE for Online Explanation on Data Streams

Explainable Artificial Intelligence (XAI) focuses mainly on batch learni...

A Novel Incremental Clustering Technique with Concept Drift Detection

Data are being collected from various aspects of life. These data can of...

iPDP: On Partial Dependence Plots in Dynamic Modeling Scenarios

Post-hoc explanation techniques such as the well-established partial dep...

A Practical Incremental Method to Train Deep CTR Models

Deep learning models in recommender systems are usually trained in the b...

Automated Machine Learning Techniques for Data Streams

Automated machine learning techniques benefited from tremendous research...

Analysis of Drifting Features

The notion of concept drift refers to the phenomenon that the distributi...

SHAP-IQ: Unified Approximation of any-order Shapley Interactions

Predominately in explainable artificial intelligence (XAI) research, the...