Zihan Zhang

Building a hybrid AI vision at Lenovo AI Technology Centre (LATC).

My research focused on high-efficiency and privacy-preserving distributed machine learning on edge devices.

• Built a distributed ML system for training language models and CNNs across 100+ edge devices; reduced end-to-end training time and improved model accuracy.
• Reconstructed training pipeline, overlapping computation with communication; reduce device–server communication and on-device computation.
• Engineered pipeline parallelism and async aggregation to minimise idle time on heterogeneous devices; stabilised throughput under stragglers and intermittent connectivity.
• Industry collaboration with Rakuten Mobile Japan: multiple US patents and top-tier publications.

I was dedicated to designing and developing eye-tracking-based authentication on edge devices that is highly accurate and requires no additional equipment.

• Addressed GenAI-era vulnerabilities in face/voice biometrics by shifting to gaze signals; co-authored the winning CyberASAP bid and secured £92k to deliver a PoC and roadmap.
• Architected an eye-tracking authentication system end to end (capture → calibration → estimation → embedding → matching) with challenge–response and basic liveness checks.
• Collected 100+ participant sessions on phone and laptop.
• Developed and delivered a PoC at CyberASAP Demo Day.

As a research/algorithm engineer at Huawei Noah's Ark Lab, I used machine learning to optimise Huawei's supply chain.

• Delivered a distributed ML-driven optimisation system for logistics (packing + routing); increased average packing rate; reduced annual logistics cost by several million USD in production.
• Built a configurable constraint framework and a learning-based solver, delivered as a tested algorithm library with API contracts; platform team containerised and deployed it on Huawei Cloud.
• Co-organised the EMO 2021 competition: shipped a standardised Docker image, automated evaluation tool, and a reproducible baseline; open-sourced the winning solution on GitHub.
• Mentored an intern and co-authored an ACM CIKM paper on ML for large-scale optimisation.