A5035429862- Anomaly Detection Techniques and Applications
- Fault Detection and Control Systems
- Software System Performance and Reliability
- Explainable Artificial Intelligence (XAI)
- Advanced Neural Network Applications
- Adversarial Robustness in Machine Learning
- AI-based Problem Solving and Planning
- Ethics and Social Impacts of AI
- Machine Learning in Materials Science
- Risk and Safety Analysis
- Military Defense Systems Analysis
- Occupational Health and Safety Research
- Time Series Analysis and Forecasting
- Network Security and Intrusion Detection
- Scientific Computing and Data Management
- Infrastructure Resilience and Vulnerability Analysis
- Guidance and Control Systems
- Artificial Intelligence in Healthcare and Education
University of California, Davis
2020-2025
Multiple-player games involving cooperative and adversarial agents are a type of problems great practical significance. In this letter, we consider an attack-defense game with single attacker multiple defenders. The attempts to enter protected region, while the defenders attempt defend same region capture outside region. We propose distributed pursuit-defense strategy for defenders' defense against attacker. Inside bounded, convex, two-dimensional space, choose among area-decreasing,...
Modern cyber-physical systems would often fall victim to unanticipated anomalies. Humans are still required in many operations troubleshoot and respond such anomalies, those future deep space habitats. To maximize the effectiveness efficiency of anomaly response process, information provided by technologies their human operators must be epistemically accessible or explainable. This paper offers a first step towards developing explainable systems. It proposes logic, Causal Signal Temporal...
Cyberphysical systems (CPSs) are vulnerable to catastrophic fault propagation due the strong connectivity among their subsystems. This article introduces a learning-based method enable CPSs explain faults human users, facilitating effective and efficient collaborative error diagnosis.
View Video Presentation: https://doi.org/10.2514/6.2023-1828.vid The National Aeronautics and Space Administration (NASA) its international partners are beginning a new era of human exploration by sustainably returning humans to the Moon preparing for missions Mars. Due great distances these mission will travel from Earth, reliance on Earth-based support be diminished greater levels autonomy required. Recent advances in Artificial Intelligence (AI) such as Machine Learning (ML) robotics make...
Time series anomaly detection is a critical machine learning task for numerous applications, such as finance, healthcare, and industrial systems. However, even high-performed models may exhibit potential issues biases, leading to unreliable outcomes misplaced confidence. While model explanation techniques, particularly visual explanations, offer valuable insights detect by elucidating attributions of their decision, many limitations still exist -- They are primarily instance-based not...
Researchers have proposed various methods for visually interpreting the Convolutional Neural Network (CNN) via saliency maps, which include Class-Activation-Map (CAM) based approaches as a leading family. However, in terms of internal design logic, existing CAM-based often overlook causal perspective that answers core "why" question to help humans understand explanation. Additionally, current CNN explanations lack consideration both necessity and sufficiency, two complementary sides...