Shapley Values
PulseAugur coverage of Shapley Values — every cluster mentioning Shapley Values across labs, papers, and developer communities, ranked by signal.
10 day(s) with sentiment data
-
ShapleyOG method accelerates Shapley value estimation in machine learning
Researchers have developed ShapleyOG, a novel method for estimating Shapley values in machine learning that significantly reduces computational overhead. By focusing on the odd component of the game, ShapleyOG requires …
-
New study compares AI explanation methods for relational databases
Researchers have developed a new methodology to compare different importance attribution methods for deep learning models operating on relational databases. Their study analyzed how these methods rank and score attribut…
-
New TempShap framework offers detailed feature attribution for time-series forecasting
Researchers have introduced TempShap, a novel framework for explaining time-series forecasting models. Unlike existing methods that treat features as single entities, TempShap analyzes individual input cells across the …
-
New research reveals flaws in AI vehicle trajectory prediction models
A new research paper published on arXiv explores the critical flaw in current vehicle trajectory prediction models, where information from surrounding agents can paradoxically degrade prediction accuracy. Researchers Da…
-
New method precisely attributes alarm scores with scarce data
Researchers have developed a new method for precisely attributing alarm scores in industrial and scientific systems, even when incident data is scarce. This approach models the alarm score change as a Shapley game, dist…
-
TRIP-Shapley framework improves P2P federated learning attribution
Researchers have developed TRIP-Shapley, a new framework for attributing contributions in peer-to-peer federated learning (P2PFL). This method addresses communication overhead and misreporting attack vulnerabilities inh…
-
New AI model forecasts tourist arrivals using social influence weighting
A new research paper proposes a hybrid approach for forecasting tourist arrivals, integrating Internet big data with a game-theoretic algorithm to weight social media opinion leaders. The method uses Empirical Wavelet T…
-
New framework audits AI fairness across multiple sensitive attributes
Researchers have developed a new framework for auditing fairness in AI models, focusing on "multi-sensitive distributional fairness." This approach uses one-dimensional Wasserstein geometry to analyze disparities across…
-
New PRIME method enhances variable importance measurement in ML
Researchers have introduced a new method called proportional inclusive marginal effects (PRIME) to measure variable importance in machine learning. PRIME addresses limitations of previous methods like Shapley-based meas…
-
New methods offer provably accurate Shapley value estimation for AI explainability · 2 sources tracked
Two new arXiv papers introduce advanced methods for estimating Shapley values, a key technique in explainable AI for attributing model predictions to input features. The first paper, "Provably Accurate Shapley Value Est…
-
New method quantifies robustness of Shapley values in tree-based models
Researchers have developed a new method using an imprecise Dirichlet model (IDM) to analyze the robustness of Shapley values in decision trees and random forests. This approach quantifies and examines interval-valued Sh…
-
New geometry links Naive Bayes classifiers to Shapley values
This paper introduces a novel geometric framework for understanding weighted Naive Bayes classifiers, particularly in the context of data streams with forgetting. Researchers developed a discriminative reformulation bas…
-
New benchmark evaluates online adaptation for EEG brain-computer interfaces
Researchers have developed a new benchmark to evaluate online adaptation techniques for electroencephalography (EEG) brain-computer interface (BCI) models. The study compared two pipeline families, Common Spatial Patter…
-
New CHOQOLATE layer organizes concept bottleneck models for better interpretability
Researchers have developed CHOQOLATE, a novel interpretable layer for Concept Bottleneck Models (CBMs) that utilizes 2-additive Choquet integrals to organize latent spaces. This method addresses the issue of entangled c…
-
New LandscapeSHAP method assigns credit in topological data analysis
Researchers have developed LandscapeSHAP, a novel method for attributing machine learning model predictions to specific features derived from topological data analysis. This approach specifically addresses the challenge…
-
New framework optimizes federated learning for healthcare centers
Researchers have developed Adaptive Bayesian Partner Selection (ABPS), a peer-to-peer framework designed to improve federated learning in healthcare settings. This approach addresses challenges like data heterogeneity a…
-
New research tackles Shapley value estimation challenges in ML · 2 papers
Two new research papers published on arXiv propose novel methods for estimating Shapley values, a key technique in machine learning for feature attribution. The first paper introduces FUSHAP, designed to handle multi-si…
-
Deep learning and explainability uncover market dynamics
Researchers have developed a novel approach to understanding complex financial market dynamics by combining deep learning with explainability techniques. By training a deep feed-forward network on high-frequency trading…
-
New contrastive explanations for AI argumentation frameworks unveiled
Researchers have introduced contrastive explanations for Quantitative Bipolar Argumentation Frameworks (QBAFs), a method for explaining the differences between two arguments rather than just one. This new approach defin…
-
AgentWorld simulation framework evaluates AI agent reliability with personality and adversarial testing
Researchers have introduced AgentWorld, a novel simulation framework designed to evaluate the reliability of agentic information retrieval systems. This framework incorporates personality-driven user populations based o…