REVIEW 13 cited by
InterpretML: A Unified Framework for Machine Learning Interpretability
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
InterpretML is an open-source Python package which exposes machine learning interpretability algorithms to practitioners and researchers. InterpretML exposes two types of interpretability - glassbox models, which are machine learning models designed for interpretability (ex: linear models, rule lists, generalized additive models), and blackbox explainability techniques for explaining existing systems (ex: Partial Dependence, LIME). The package enables practitioners to easily compare interpretability algorithms by exposing multiple methods under a unified API, and by having a built-in, extensible visualization platform. InterpretML also includes the first implementation of the Explainable Boosting Machine, a powerful, interpretable, glassbox model that can be as accurate as many blackbox models. The MIT licensed source code can be downloaded from github.com/microsoft/interpret.
Forward citations
Cited by 13 Pith papers
-
From population norms to personalized trajectories: interpretable Bayesian forecasting for cognitive decline
A personalized Bayesian monitoring framework, built from an interpretable baseline model plus visit-to-visit probability updating, flagged emerging cognitive decline before study-defined worsening in 31% of HRS and 41...
-
The Quadrilateral Loss: Additivity as a Measurable Behavior of Dense Neural Networks
A finite-difference penalty makes neural-network additivity measurable and tunable, and route comparisons show interaction magnitude is a weak predictor of what regularization retains.
-
Geometry-Aware Uncertainty Coresets for Robust Visual In-Context Learning in Histopathology
GAUC selects histopathology ICL coresets in VLM embedding space by jointly optimizing MMD fidelity, prompt-robust mutual-information regularization, and entropy-based uncertainty, matching baseline accuracy with bette...
-
Interpretable, Physics-Informed Learning Reveals Sulfur Adsorption and Poisoning Mechanisms in 13-Atom Icosahedra Nanoclusters
Across 30 transition metals at the fixed icosahedral motif, sulfur adsorption is driven mainly by the metal–sulfur interaction with mostly modest distortion; Ti13, Zr13, Hf13 emerge as the balanced, structurally resil...
-
Long-Term Variability in Physiological-Arousal Relationships for Robust Emotion Estimation
The mapping from physiological signals to subjective arousal shifts within individuals over months, and emotion estimation models must be periodically updated to stay accurate.
-
CoFrNets: Interpretable Neural Architecture Inspired by Continued Fractions
CoFrNet represents functions as stacked continued fractions with linear layers and reciprocal activation, but the universal approximation proof applies to a more general function class than the architecture used in ex...
-
A Human-Centric Approach to Explainable AI for Personalized Education
This thesis shows that post-hoc explainers disagree systematically in educational models, and that intrinsically interpretable architectures (MultiModN, InterpretCC) with an LLM explanation layer (iLLuMinaTE) keep acc...
-
FAIR_XAI: Improving Multimodal Foundation Model Fairness via Explainability for Wellbeing Assessment
Zero-shot vision-language models are unreliable and vary widely for depression screening, and explainability-based fairness interventions often trade away accuracy without reliable fairness gains.
-
Transparent and Fair Profiling in Employment Services: Evidence from Switzerland
On Swiss employment data, explainable boosting machines predict long-term unemployment almost as accurately as XGBoost while remaining fully interpretable.
-
midr: Learning from Black-Box Models by Maximum Interpretation Decomposition
The midr package builds a global additive surrogate for any black-box model by least-squares projection with strict centering constraints, yielding interpretable main effects, interactions, and SHAP-style attributions.
-
Economic Complexity as a Determinant of Regional Human Development in Brazil: Evidence across Aggregation Scales
At Brazil’s Immediate Region scale, adapted economic complexity alone largely predicts IDHM (EBM R²≈0.82 with road-network metrics); municipal models are noisier and dominated by a Northeast dummy.
-
Unveiling Location-Specific Price Drivers: A Two-Stage Cluster Analysis for Interpretable House Price Predictions
Two-stage clustering before linear regression or an additive model improves German house-price prediction accuracy relative to global models on the 2023 RED dataset, though the reported improvement percentages are int...
-
ExplainBench: A Benchmark Framework for Local Model Explanations in Fairness-Critical Applications
ExplainBench is a proposed open-source benchmarking suite for local explanation methods, but the paper provides no empirical evaluation.
Discussion (0). Sign in to comment.