Muhammad Rashid

Machine Learning Researcher specializing in Explainable AI (XAI),
Computer Vision, and Visual Anomaly Detection for real-world systems.

🔬 Research Focus

• Explainable AI for trustworthy decision-making
• Visual anomaly detection in industrial environments
• Pixel-level feature attribution (Shapley-based methods)

📌 AAAI 2026: ShapBPT — Image Feature Attribution using Data-Aware Binary Partition Trees

⚙️ Systems & Engineering

• Real-time anomaly detection for robotics safety (ROS2-based pipeline)
• VAE / VAE-GAN models for industrial anomaly detection
• Multi-camera monitoring and safety area segmentation
• End-to-end pipelines: training → calibration → live inference

🚀 Achieved:

• ~12 FPS real-time inference
• ~99.6% detection accuracy in industrial scenarios

🧠 Open Source & Tools

• ShapBPT (AAAI 2026) — Explainability framework
• LIME Stratified Sampling — Improved explanation stability
• AI on Edge Devices — Lightweight deployment pipelines
• XAI evaluation tools (AUC, saliency metrics, benchmarking)

📄 Selected Publications

• ShapBPT: Image Feature Attribution using Data-Aware Binary Partition Trees
  AAAI 2026

• Using Stratified Sampling to Improve LIME Image Explanations
  XAI / Explainability

• Can I Trust My Anomaly Detection System?
  Explainable anomaly detection case study

➡️ See full publication list

🚀 Featured Projects

➡️ View all projects

🔧 Real-Time Anomaly Detection for Robotics

• Industrial safety monitoring system
• Multi-area detection (RoboArm, Conveyor Belt, etc.)
• ROS2 integration + live inference + anomaly maps

🧠 ShapBPT (AAAI 2026)

• Novel explainability method for images
• Uses hierarchical Binary Partition Trees
• More data-aware than standard SHAP partition explainer

📊 LIME Stratified Sampling

• Improves stability and reliability of explanations
• Reduces variance in perturbation-based methods

🌍 Experience

• 🎓 PhD Researcher — University of Turin (Italy)
• 🏭 Industrial Research — RuleX Innovation Labs
• 🇪🇺 EU Project — DistriMuSe (Robotics Safety & AI)

🤝 Collaboration

I’m interested in collaborations on:

• Explainable AI (XAI)
• Trustworthy anomaly detection
• Industrial AI systems
• Vision-language models (emerging interest)

📫 Contact

• 📧 Email
• 🔗 GitHub
• 📚 Google Scholar