Reproduce “Transfer Learning Assisted XgBoost For Adaptable Cyberattack Detection In Battery Packs” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Sanchita Ghosh, Tanushree Roy
Researchers at Texas Tech University developed a Transfer Learning-assisted XgBoost (TL-XGB) algorithm for detecting cyberattacks on electric vehicle battery pack sensor data. The method accurately identifies data swap and replay attacks in simulated large-format battery packs, demonstrating adaptability and computational efficiency with minimal pack-level training data.
Key insights
- Transferring knowledge from abundant cell-level data to scarce pack-level data significantly reduces the need for extensive data when adapting models to new battery configurations.
- The combination of Transfer Learning with XgBoost offers a computationally efficient and adaptable solution for real-time cyberattack detection in battery systems.
- A residual-based detection strategy, which does not require pre-training with attacked data, enhances the algorithm's robustness against various and unforeseen cyberattack patterns.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.