Model Agnostic Supervised Local Explanations

By Gregory Plumb, Denali Molitor, Ameet Talwalkar. 2018

224 citations recorded by the source corpus.

Model interpretability is an increasingly important component of practical machine learning. Some of the most common forms of interpretability systems are example-based, local, and global explanations. One of the main challenges in interpretability is designing explanation systems that can capture aspects of each of these explanation types, in order to develop a more thorough understanding of the model. We address this challenge in a novel model called MAPLE that uses local linear modeling techniques along with a dual interpretation of random forests (both as a supervised neighborhood approach and as a feature selection method). MAPLE has two fundamental advantages over existing interpretability systems. First, while it is effective as a black-box explanation system, MAPLE itself is a highly accurate predictive model that provides faithful self explanations, and thus sidesteps the typical accuracy-interpretability trade-off. Specifically, we demonstrate, on several UCI datasets, that MAPLE is at least as accurate as random forests and that it produces more faithful local explanations than LIME, a popular interpretability system. Second, MAPLE provides both example-based and local explanations and can detect global patterns, which allows it to diagnose limitations in its local explanations.

Subfields: Artificial Intelligence, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Control and Systems Engineering

Citations in the corpus

Regularizing Black-box Models for Improved Interpretability (2019, methodology): “The ﬁrst, ExpO-Fidelity , is designed for semantic features and explainers that directly make predictions, such as (Ribeiro et al., 2016; Lundberg and Lee, 2017; Plumb et al., 2018).”
On the (In)fidelity and Sensitivity for Explanations. (2019, background): “We note that our deﬁnition of local and global explanations follows the description of [3], distinct from that in [22].”
Benchmarking and survey of explanation methods for black box models (2021, methodology): “MAPLE (Plumb et al. 2018) is a local post-hoc model-agnostic explainer that can also be used as a transparent model due to its internal structure.”
A Study of Automatic Metrics for the Evaluation of Natural Language Explanations (2021, background): “, 2016); MAPLE can provide feedback for all three types of explanations: example-based, local and global explanations (Plumb et al., 2018); CLEAR explains a single prediction by using local explanations that include statements of key counterfactual cases (White and d’Avila Garcez, 2019).”
FACE: Feasible and Actionable Counterfactual Explanations (2019, unknown): “It imposes no limitations on the complexity of the compatible models, nevertheless popular approaches in XML, such as “Global Surrogate Models” or “Local Surrogate Models” [12, 13, 11], add an extra layer of modelling complexity in exchange for transparency.”
Explainable AI for Trees: From Local Explanations to Global Understanding (2019, methodology): “This can be done by training a global mimic model to approximate the original model, then locally explaining the mimic model by either taking its gradient [4], or fitting a local linear model as in MAPLE [53].”
Exchange-correlation functionals for band gaps of solids: benchmark, reparametrization and machine learning (2020, unknown): “(ii) The model agnostic supervised local explanations (called MAPLE) 146 is a recently developed machine-learning technique that combines the usability and accuracy of decision trees with the interpretability of local linear models and self-explanations, therefore circumventing the usual…”
A Survey on Neural Network Interpretability (2020, methodology): “MAPLE [69] is a similar method using local linear models as explanations.”
A Survey on the Explainability of Supervised Machine Learning (2020, methodology): “Model Agnostic Supervised Local Explanations (MAPLE) (Plumb et al., 2018) uses a local linear modeling approach with a dual interpretation strategy of random forests.”
Deterministic Local Interpretable Model-Agnostic Explanations for Stable Explainability (2021, unknown): “These approaches can be grouped based on different criterion [1–4] Among them, post hoc model agnostic approaches are popular in practice, where the target is to design a separate algorithm that can explain the decision making process of any ML model.”
RL-LIM: Reinforcement Learning-based Locally Interpretable Modeling (2019, methodology): “Model Agnostic Supervised Local Explanations (MAPLE) (Plumb et al., 2018) furthers adds a method for feature selection on top of SILO – it utilizes ad-hoc tree-based ensemble methods to determine the weights of training instances for each target instance and uses the weights to opti-2 mize a…”
Interpretable deep learning: interpretation, interpretability, trustworthiness, and beyond (2021, methodology): “Similarly, several model-agnostic algorithms, such as Anchors [138], SHAP [110], RISE [127],MAPLE [130], target interpreting features and provide feature importance or contributions to the final decision.”
Evaluating and Aggregating Feature-based Model Explanations (2020, background): “High faithfulness appears as relevance in [ Samek et al. , 2016 ] , as gold set in [ Ribeiro et al. , 2016 ] , as faithfulness in [ Plumb et al. , 2018 ] , as sensitivity-n in [ Ancona et al. , 2018 ] , and as inﬁdelity in [ Yeh et al. , 2019 ] .”