Get Started
Research questionHow can we explain interacting reasons behind classifier predictions fast enough for repeated local use?Feature-attribution scores can obscure combinations of features that jointly influence a prediction. Generating detailed local explanations repeatedly can also be too costly for latency-sensitive applications.
Inference Optimization
Machine Learning
Statistical Machine Learning
Latest papersRecent research connected to this question, newest first.ProToMEx: Rapid, Interpretable Explanations via Structured RepresentationsThe source addresses model-agnostic explanations for classifiers, with global and local explanations evaluated on standardized tabular and synthetic datasets. It reports fidelity comparable to SHAP and LIME and lower amortized cost for local explanations; the supplied evidence does not establish performance across other data modalities or deployment settings.research paper · Sep 2, 2026
Related questions
How can interval-based time-series classifiers speed up frequent inference without materially reducing accuracy?How can differentially private federated learning preserve faithful explanations without sacrificing predictive utility?How can we generate realistic, minimally edited visual counterfactuals without inheriting classifier shortcuts?How can teams estimate whether multiclass accuracy gains reduce optimization regret before building a model?
Home
Topics
Search
Library