arXiv research

The problem of explaining deep learning models, and model predictions generally, has attracted intensive interest recently. Many successful approaches forgo global approximations in order to provide more faithful local interpretations of the model's behavior. LIME develops multiple interpretable models, each approximat…

Develops transparent global models consistent with local explanations.

Proposes a framework to explain complex global forecasting models.

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 th…

Proposes DR-ME test for interpretable distributional treatment effects.

The paper introduces closed-form expressions for interpreting Tsetlin Machines.

A method interprets black-box models using an ensemble of gradient boosting machines.

Locally adaptive nearest neighbors improve automated systems' performance and are easier to interpret.

Understanding black-box machine learning models is crucial for their widespread adoption. Learning globally interpretable models is one approach, but achieving high performance with them is challenging. An alternative approach is to explain individual predictions using locally interpretable models. For locally interpre…

Improved local feature attributions using neighbourhood reference distributions.

Proposes a model to interpret complex ML algorithms.

Unified method for local GBDT feature contributions.

Improved local explainer aggregation for interpretable machine learning models.

MaGNet integrates local and global graph information for interpretable results.

New approach combines PCA and t-sne for better data analysis.

LI-ITR combines flexible ML with interpretable approximations for personalized treatment rules.

A framework uses preprocessing to improve psychiatric questionnaire predictions while maintaining interpretability.

We introduce the localized Lasso, which is suited for learning models that are both interpretable and have a high predictive power in problems with high dimensionality $d$ and small sample size $n$. More specifically, we consider a function defined by local sparse models, one at each data point. We introduce sample-wis…

Local MDI+ improves feature importance for tree-based models, enhancing interpretability and performance.

New method identifies important features and interactions in RF models.

HALO uses local Lipschitz constants to optimize functions efficiently.

Model-agnostic interpretation methods can mislead if not used carefully.

Study robustness of global feature effect explanations in machine learning models.

PerCDL learns personalized dictionaries for physiological signals combining global and local structures.

A barrier to the wider adoption of neural networks is their lack of interpretability. While local explanation methods exist for one prediction, most global attributions still reduce neural network decisions to a single set of features. In response, we present an approach for generating global attributions called GAM, w…

We provide a complete solution to the problem of extending a local Lie groupoid to a global Lie groupoid. First, we show that the classical Mal'cev's theorem, which characterizes local Lie groups that can be extended to global Lie groups, also holds in the groupoid setting. Next, we describe a construction that can be …

New methods compare local and global feature importance scores for bioinformatics models.

Tree-based machine learning models such as random forests, decision trees, and gradient boosted trees are the most popular non-linear predictive models used in practice today, yet comparatively little attention has been paid to explaining their predictions. Here we significantly improve the interpretability of tree-bas…

Deep learning models designed for visual classification tasks on natural images have become prevalent in medical image analysis. However, medical images differ from typical natural images in many ways, such as significantly higher resolutions and smaller regions of interest. Moreover, both the global structure and loca…

DNN2LR bridges DNN power and LR interpretability.

New method visualizes tabular feature semantics for better model understanding.

Bundling of graph edges (node-to-node connections) is a common technique to enhance visibility of overall trends in the edge structure of a large graph layout, and a large variety of bundling algorithms have been proposed. However, with strong bundling, it becomes hard to identify origins and destinations of individual…

GAMLA learns manifold structures with auto-encoding for global insights.

Study compares and contrasts various ML explanation methods, highlighting their disagreements and similarities.

Paper proposes using tree-based surrogate models for efficient Shapley computation.

Supervised Machine Learning (SML) algorithms such as Gradient Boosting, Random Forest, and Neural Networks have become popular in recent years due to their increased predictive performance over traditional statistical methods. This is especially true with large data sets (millions or more observations and hundreds to t…

TopoGeoScore selects robust checkpoints using only source-domain representations.

GADGET framework decomposes global feature effects using recursive partitioning.

In an earlier work we identified the types and numbers of static equilibrium points of solids arising from fine, equidistant $n$-discretrizations of smooth, convex surfaces. We showed that such discretizations carry equilibrium points on two scales: the local scale corresponds to the discretization, the global scale to…

Interpretable surrogates of black-box predictors trained on high-dimensional tabular datasets can struggle to generate comprehensible explanations in the presence of correlated variables. We propose a model-agnostic interpretable surrogate that provides global and local explanations of black-box classifiers to address …

Pointwise localization allows more precise localization and accurate interpretability, compared to bounding box, in applications where objects are highly unstructured such as in medical domain. In this work, we focus on weakly supervised localization (WSL) where a model is trained to classify an image and localize regi…

Unified approach for interpretable regression with flexible modeling.

Global neural networks improve financial forecasting accuracy with larger, diverse datasets.

The main new result here is the cancellation of global anomalies in the Type I superstring, with and without D-branes. Our argument here depends on a precise interpretation of the 2-form abelian gauge field using KO-theory; then the anomaly cancellation follows from a geometric form of the full Atiyah-Singer index theo…

We study local, global and local-to-global properties of threefolds with certain singularities. We prove criteria for these threefolds to be rational homology manifolds and conditions for threefolds to satisfy rational Poincaré duality. We relate the topological Euler characteristic of elliptic Calabi-Yau threefolds wi…

XDeep is an open-source Python package developed to interpret deep models for both practitioners and researchers. Overall, XDeep takes a trained deep neural network (DNN) as the input, and generates relevant interpretations as the output with the post-hoc manner. From the functionality perspective, XDeep integrates a w…

Study interprets deep learning models for Heston model in finance.

Interactions such as double negation in sentences and scene interactions in images are common forms of complex dependencies captured by state-of-the-art machine learning models. We propose Mahé, a novel approach to provide Model-agnostic hierarchical éxplanations of how powerful machine learning models, such as deep ne…