ViTaX provides formal guarantees for targeted explanations in safety-critical systems.
problem Need trustworthy explanations for safety-critical deep neural networks.
method Formal reachability analysis for targeted, semifactual explanations.
result First method to provide formally guaranteed explanations of model resilience.
Proposes a method for clearer counterfactual explanations of deep networks.
problem Unclear explanations from current counterfactual methods.
method Gradual construction of explanations through masking and composition steps.
result Produces human-friendly, interpretable explanations with fewer modifications.
Paper introduces new evaluation criteria for feature-based model explanations.
problem Establishing reliable feature importance explanations for models.
method Robustness analysis using smaller adversarial perturbations.
result New explanations that are necessary and sufficient for predictions.
XAI methods fail to explain ML models reliably.
problem Current XAI methods fail to provide reliable explanations for ML models.
method Formally define problems and design methods accordingly.
result Diverse notions of explanation correctness and metrics needed.
Many proposed methods for explaining machine learning predictions are in fact challenging to understand for nontechnical consumers. This paper builds upon an alternative consumer-driven approach called TED that asks for explanations to be provided in training data, along with target labels. Using semi-synthetic data fr…
PWSHAP provides targeted explanations for complex models.
problem Inability of black-box models to explain targeted effects in sensitive domains.
method Augments model with DAG, uses Shapley values for causal pathway identification.
result Establishes error bounds and demonstrates resolution, interpretability, and locality.
Transfer learning which aims at utilizing knowledge learned from one problem (source domain) to solve another different but related problem (target domain) has attracted wide research attentions. However, the current transfer learning methods are mostly uninterpretable, especially to people without ML expertise. In thi…
New approach to meaningful and robust algorithmic recourse.
problem Ineffective and unmeaningful algorithmic recourse explanations.
method Meaningful Algorithmic Recourse (MAR) and Effective Algorithmic Recourse (EAR).
result Proposes new constraints for algorithmic recourse that improve both prediction and target.
Attack reveals model details from counterfactual explanations.
problem Extracting model details from counterfactual explanations.
method Adversary uses counterfactual explanations to build high-fidelity model.
result High-fidelity and high-accuracy model extraction possible.
LLMs' explanations are often insufficient and vary with input distribution.
problem Evaluating the sufficiency of LLM explanations without predefined biases.
method Generalizing sufficiency to arbitrary explanations, using LLM's input beliefs, and introducing SCSuff metric.
result Explanation sufficiency can vary with input distribution and is weakly correlated with model size, accuracy, or output entropy.
New method provides calibrated feature importance explanations for regression models.
problem Lack of uncertainty quantification in existing local explanation methods.
method Extension of Calibrated Explanations method to support regression and probabilistic regression.
result Calibrated Explanations for regression provides quantified uncertainty and robust explanations.
This research integrates attention into XAI frameworks for better model explanations.
problem Improving the interpretability of transformer models.
method Developed two novel explanation methods: Shapley value decomposition and token-level directional derivatives.
result Attention weights can be meaningfully incorporated into XAI frameworks, enhancing transformer explainability.
Generative models explain machine learning predictions with counterfactual instances.
problem Generating human-interpretable insights into machine learning model predictions.
method Sparse counterfactual explanations using conditional generative models.
result Single forward pass generates batches of counterfactual instances.
OrphicX generates causal explanations for GNNs by isolating latent causal factors.
problem Generating interpretable causal explanations for complex graph neural networks.
method Develops a generative model and objective function to isolate latent causal factors, maximizing information flow.
result OrphicX effectively identifies causal semantics, significantly outperforming alternatives.
Proposes a method to explain black-box models using causal learning.
problem Existing explainability methods focus on micro-level inputs, not interpretable features.
method Learns causal graphical representations to differentiate between causal and confounding influences.
result Graphs can differentiate between interpretable and confounding features.
New approach models how explanations shift with distribution changes.
problem Model performance drops with changing input data distributions.
method Models explanation shifts and compares them to state-of-the-art techniques.
result Modeling explanation shifts better detects out-of-distribution behavior.
New method improves model reconstruction using counterfactuals and polytope theory.
problem Reconstructing models with minimal input changes and avoiding decision boundary shifts.
method Using polytope theory to derive loss functions that treat counterfactuals differently from ordinary instances.
result Improves fidelity between target and surrogate model predictions on multiple datasets.
Improved interpretability methods for ML models using local regressions and variable importance.
problem Inability of existing interpretability methods to provide reliable explanations for ML models, especially in high-dimensional problems with irrelevant features and non-linear relationships.
method Introduces VarImp and SupClus methods using local regressions with weighted distance considering variable importance.
result VarImp and SupClus methods yield better explanations than state-of-the-art approaches, especially in high-dimensional problems with irrelevant features and non-linear relationships.
HSIC-based method explains GNN structures.
problem Interpreting GNNs' complex dynamics.
method Flexible model-agnostic explanation using HSIC.
result Identifies crucial structures in GNNs.
New method uses SHapley Additive Explanations to identify anomaly detectors with complementary behaviors.
problem Challenges in unsupervised anomaly detection due to diverse data distributions and lack of labels.
method Characterize anomaly detectors using SHapley Additive Explanations to measure feature importance and similarity.
result Detectors with similar explanations produce correlated anomaly scores, while those with divergent explanations are complementary.
New method generates plausible counterfactuals for time series classification.
problem Generating realistic counterfactuals for time series data.
method Gradient-based optimization with soft-DTW alignment and multi-faceted loss function.
result Our method outperforms existing approaches in temporal realism and distributional alignment.
RbX generates local explanations for black-box models using greedy polytope construction.
problem Generating interpretable local explanations for complex models.
method Greedy algorithm for building a convex polytope to approximate feature regions.
result RbX can more accurately identify important features compared to existing methods.
This paper introduces DCE for better counterfactual explanations using optimal transport.
problem Lack of nuanced distributional characteristics in existing counterfactual explanations.
method Formulates a chance-constrained optimization problem using optimal transport to derive counterfactual distributions.
result DCE provides deeper insights into decision-making models by aligning counterfactual distributions with factual ones.
Generative adversarial approach for satellite image time series land cover classification.
problem Enhance interpretability of land cover classification models.
method Generative adversarial counterfactual approach for multi-class land cover classification.
result Discovery of interesting information on land cover class relationships and sparser, interpretable solutions.
The concept of explainability is envisioned to satisfy society's demands for transparency on machine learning decisions. The concept is simple: like humans, algorithms should explain the rationale behind their decisions so that their fairness can be assessed. While this approach is promising in a local context (e.g. to…
PatternLocal improves XAI for non-linear models by suppressing suppressor variables.
problem Suppressor variables cause false-positive feature attributions in non-linear models.
method PatternLocal uses locally linear surrogate models and transforms weights into a generative representation.
result PatternLocal reduces false-positive attributions and provides more reliable explanations.
Proposes counterfactual explanations for deep two-sample tests on high-dimensional data.
problem Limited interpretability of deep two-sample tests on high-dimensional data.
method Combines diffusion autoencoder and pretrained deep two-sample test model to generate counterfactuals.
result Counterfactual transformations increase p-values, indicating closer distribution similarity.
We focus on the problem of search in the multilingual setting. Examining the problems of next-sentence prediction and inverse cloze, we show that at large scale, instance-based transfer learning is surprisingly effective in the multilingual setting, leading to positive transfer on all of the 35 target languages and two…
Proposes GPR with local explanation for feature interpretation.
problem Lack of feature interpretation in Gaussian process regression.
method Introduces a locally linear model to explain feature contributions in GPR predictions.
result Achieves comparable predictive performance to GPR and superior interpretability.
SurvFD and SurvSHAP-IQ provide interpretable survival models by analyzing feature interactions.
problem Non-additivity of hazard and survival functions limits standard additive explanation methods.
method SurvFD decomposes higher-order effects into time-dependent and time-independent components, extending Shapley interactions to time-indexed functions.
result SurvFD and SurvSHAP-IQ offer a new perspective on survival explanations, explicitly characterizing feature interactions.
The need for transparency of predictive systems based on Machine Learning algorithms arises as a consequence of their ever-increasing proliferation in the industry. Whenever black-box algorithmic predictions influence human affairs, the inner workings of these algorithms should be scrutinised and their decisions explai…
Survey of counterfactual explanations for time series classification.
problem Generating plausible and actionable counterfactuals for time series data.
method Review of various counterfactual generation methods for time series classification.
result Highlight unique challenges and strengths of existing methods.
Equivariant localization techniques give a rigorous interpretation of the Witten genus as an integral over the double loop space. This provides a geometric explanation for its modularity properties. It also reveals an interplay between the geometry of double loop spaces and complex analytic elliptic cohomology. In part…
The study explains how neural networks align their kernels to target functions during training.
problem Understanding how neural networks align their kernels to target functions during training.
method Theoretical analysis of kernel evolution in toy models and deep networks.
result Kernel alignment naturally emerges during training to accelerate convergence and improve generalization.
Sharp theory of neural network scaling laws for hierarchical targets.
problem Learning hierarchical multi-index models in neural networks.
method Sharp information-theoretic scaling laws derived for two-layer neural networks.
result Optimal rates achieved by a simple spectral estimator.
Proposes a new method to explain model predictions for consumer recourse.
problem Current explanation methods fail to provide meaningful recourse to decision subjects.
method Develops feature responsiveness scores to highlight actionable features.
result Standard practices can undermine decision subjects by highlighting unresponsive features.
Feature importance estimates that inform users about the degree to which given inputs influence the output of a predictive model are crucial for understanding, validating, and interpreting machine-learning models. However, providing fast and accurate estimates of feature importance for high-dimensional data, and quanti…
This work analyzes how frequency components affect CNN predictions and robustness.
problem Lack of frequency-based explanation for CNNs leading to vulnerabilities.
method Frequency component analysis and quantification of their contribution to CNN predictions.
result Adversarial attacks exploit high-frequency features, while robustness comes from low-frequency associations.
This paper analyzes how periodic and soft target updates stabilize linear Q-learning.
problem Theoretical explanation of stabilization mechanisms for linear Q-learning.
method Exact analysis using switched linear system dynamics and the joint spectral radius.
result Periodic and soft target updates can guarantee convergence to the exact projected Q-Bellman solution under specific conditions.
A deep RL approach generates counterfactual instances efficiently.
problem Efficient generation of counterfactual instances for large datasets and diverse models.
method Deep reinforcement learning to optimize counterfactual instances in a single forward pass.
result Model-agnostic and scalable counterfactual generation.
The paper develops statistical inference methods for SHAP values.
problem Lack of statistical inference for SHAP values in model-agnostic feature importance.
method Semi-parametric approach using U-statistics and Neyman orthogonal scores for functionals of nested regressions.
result Asymptotically normal estimates of the pth powers of SHAP values for various p.
CONFETTI improves interpretability of deep learning models for MTS by providing counterfactual explanations.
problem Lack of transparency in deep learning models for multivariate time series classification.
method CONFETTI is a novel multi-objective counterfactual explanation method that balances prediction confidence, proximity, and sparsity.
result CONFETTI outperforms state-of-the-art methods in various metrics, improving interpretability and decision support.
Unified framework for distribution shift estimation, explanation, and improvement.
problem Estimating, explaining, and improving model performance on target domains with distribution shift.
method Entropic Projection Alignment (EPA) aligns source and target distributions by matching moments and minimizing KL divergence.
result EPA consistently outperforms state-of-the-art baselines while offering computational efficiency.
Many methods to explain black-box models, whether local or global, are additive. In this paper, we study global additive explanations for non-additive models, focusing on four explanation methods: partial dependence, Shapley explanations adapted to a global setting, distilled additive explanations, and gradient-based e…
AXE evaluates explanations to avoid misleading Rashomon set model selection.
problem Evaluating explanations for Rashomon set models to avoid false selection.
method Proposed AXE method to evaluate explanation quality.
result AXE detects adversarial fairwashing with 100% success rate.
The paper proposes criteria and methods for evaluating and aggregating feature-based model explanations.
problem Lack of quantitative evaluation criteria for feature-based model explanations.
method Developed quantitative evaluation criteria (low sensitivity, high faithfulness, low complexity), devised a framework for aggregation, and derived a new aggregate Shapley value explanation function.
result A new aggregate Shapley value explanation function that minimizes sensitivity.
New definition reveals encoding explanations that retain predictive power.
problem Challenges in evaluating and identifying encoding explanations.
method Developed a definition of encoding based on conditional dependence.
result Existing evaluation scores do not rank non-encoding explanations correctly, but STRIPE-X does.
This work describes an outlier detection procedure (named "OutlierTree") loosely based on the GritBot software developed by RuleQuest research, which works by evaluating and following supervised decision tree splits on variables, in whose branches 1-d confidence intervals are constructed for the target variable and pot…