This research addresses the challenge of explainability in AI for financial fraud detection, crucial for U.S. regulatory compliance. It introduces the SHAP-Guided Adaptive Ensemble (SGAE) method, which dynamically adjusts ensemble weights based on SHAP attribution agreement, achieving high performance and transparency.
An XGBoost classifier was developed using clinical features from the ADNI dataset for multi-class detection of normal cognition, mild cognitive impairment, and Alzheimer's disease. The model achieved a high mean macro AUC of 0.983 and an accuracy of 0.944, with SHAP values providing feature explainability.
This thesis presents a comprehensive study on explainable artificial intelligence planning (XAIP) for hybrid systems, addressing the growing need to generate explanations for AI-based systems. It focuses on how automated planning is applied to complex and safety-critical domains like smart energy grids and self-driving cars.
This article details a developer's epiphany while debugging a black-box Reinforcement Learning agent failing to synchronize smart agriculture microgrids. The realization that the agent lacked causal understanding led to exploring Explainable AI and causal inference frameworks to prevent cascading power failures.
Este conteúdo descreve o Topaz, um framework para roteamento auditável em fluxos de trabalho de agentes de IA. Ele visa resolver a falta de transparência na seleção de modelos, que atualmente prioriza custo e desempenho sem registrar as compensações subjacentes, utilizando perfis de habilidades e algoritmos de roteamento rastreáveis.
Blaze Balance Engine SaaS is an AI-guided system designed for monitoring, forecasting, explainability, and operational control. It features elements like live state mapping and explainable decision receipts, developed and proven in a high-activity live environment before being offered as a SaaS product.
The ST-GAT framework provides an explainable Graph Neural Network solution for early detection of bank distress and interbank contagion surveillance in the U.S. banking sector. It models over 8,000 FDIC institutions using dynamic graphs, achieving high performance (AUPRC 0.939) and identifying key predictive factors like ROA and NPL Ratio.
The article argues that SHAP is not production-ready due to issues like slowness, inconsistency, and being disconnected from the main model. The author criticizes the separate explainer architecture and proposes an approach where explanations are generated alongside the model's inference.
ReSS is a framework that bridges symbolic and neural reasoning models for tabular data prediction, aiming for both high accuracy and understandable reasoning. It leverages decision trees to extract symbolic scaffolds that guide an LLM to generate natural-language reasoning, which is then used to fine-tune specialized tabular reasoning LLMs.
Agentopic is a novel agent-based workflow for explainable topic modeling that leverages the reasoning capabilities of Large Language Models (LLMs). It enhances transparency by enabling users to trace the reasoning behind topic assignments, achieving an F1-score of 0.95, matching GPT-4.1.
The AI industry has developed powerful systems capable of automating complex workflows, but a significant challenge has emerged: the inability to fully explain their actions. This "agentic infrastructure" raises concerns about operational accountability and security, necessitating solutions like AITracer.
The author delves into the intersection of causal inference and reinforcement learning for optimizing circular manufacturing supply chains. This approach aims to develop more robust and explainable RL agents for low-power autonomous deployments.
This research focuses on overcoming the limitations of standard Reinforcement Learning models in optimizing wildfire evacuations. The author applies causal inference, inspired by Judea Pearl and Bernhard Schölkopf, to address inexplicable recommendations and confounding variables.
The text discusses the need for explainable and causal AI in space operations, illustrating with a satellite incident where an automated correction violated data sovereignty regulations. It highlights the failure of traditional AI approaches to handle the complexity of technical constraints, operational priorities, and jurisdictional boundaries.
The author built an AI news site to bridge the gap between oversimplified and academically dense AI content. It aims to provide practical, clear information for professionals who use AI tools daily.
Four students developed a production AI governance platform using MongoDB's free tier to address the critical need for explainable, traceable, and auditable AI decisions. This platform allows compliance officers to understand "why" an AI made a specific decision, a challenge intensified by rising regulatory scrutiny and GDPR penalties.
This paper provides a causal analysis of Binary Spiking Neural Networks (BSNNs) to explain their behavior, representing their spiking activity as a binary causal model. By leveraging logic-based methods like SAT and SMT solvers, it computes abductive explanations for network classifications and demonstrates that these explanations do not contain irrelevant features, unlike SHAP.
This survey explores the integration of surrogate modeling and Explainable AI (XAI) for complex system simulations, addressing the inherent black-box nature of these models. It aims to reconnect these complementary fields by outlining how XAI can unpack surrogate models despite engineering constraints.
This paper introduces ViTaX, a formal XAI framework designed to generate targeted semifactual explanations with mathematical guarantees. It addresses the shortcomings of existing XAI methods in providing trustworthy explanations for deep neural networks in safety-critical domains like autonomous driving and medical diagnosis.
This paper presents a comparative study of three explainability techniques (Integrated Gradients, Attention Rollout, and SHAP) on a fine-tuned DistilBERT model for sentiment classification. The study concludes that gradient-based attribution provides more stable and intuitive explanations, while attention-based methods are computationally efficient but less aligned with prediction-relevant features.