Biomedical Image Segmentation Exploiting Computation Power of Blockchain
• Proposed a novel consensus to exploit the computation power of blockchain for biomedical image segmentation.
• Designed augmented block header and body, task scheduler in block interval and neural network regularization for the model protection.
• Embedded watermark in deep learning models which will protect the ownership.
Real-time Adversarial Attack Target on Speech Recognition System
• Explored the intriguing properties of speech recognition system which could be critical threats in the real world scenario.
• The first to propose a real-time adversarial attack scheme for deep learning-based speech recognition models with streaming inputs.
Automatic Virtual 3D City Reconstruction
• Proposed an application orientated pipeline to generate a virtual 3D city with relative simple layout design.
• Applied Pixel2Pixel model to generate building texture from facade design which improves the quality of virtual 3D city model.
Privacy-Aware Cost-Effective Scheduling Considering Non-Schedulable Appliances
• Developed a scheduler which aims to minimize the expected sum of electricity cost under the privacy awareness constraint in a smart home system.
• Developed an iterative alternative algorithm to schedule appliances and rechargeable battery thus protects customer privacy and alleviates the influence of operation uncertainties introduced by non-schedulable appliances.
Other Projects
i) Short Video Understanding Challenge: predicted the probability that each user finishes watching and likes a given video based on a content-based, collaborative-based, and knowledge-based dataset.
ii) Hybrid movie recommendation: suggested movies based on user preference, genre, and theme, and collaborative information, respectively.
iii) Live basic math auto grader: applied for checking the handwritten basic math equations based on the MNIST classifier.
iv) Language translator: implemented a statistical machine translation model that applied finite automata, transducer, n-gram language model and hard EM algorithm.
Split-Ring Resonator(SRR) Metamaterials Simulation and Fabrication
• Designed and simulated the periodic structure with COMSOL Multiple Physics.
• Fabricated the SRR devices with Focus-Ion-Beam and evaluated with Fourier-Transform Infrared Spectroscopy
MOSFET Fundamentals, Simulation, and Measurement
• Calculated depletion region, working function, threshold voltage, MOS capacitor and MOSFET operation with equations, and simulated MOS and MOSFET with simulators (ATHENA and ATLAS Simulation (Silvaco)).
• Measured and analyzed NMOS capacitor characterization using high frequency C-V measurement.
Chenli, Changhao, Boyang Li, Yiyu Shi, and Taeho Jung. "Energy-recycling blockchain with proof-of-deep-learning." In 2019 IEEE International Conference on Blockchain and Cryptocurrency (ICBC), pp. 19-23. IEEE, 2019.
Li, Boyang, Jie Wu, and Yiyu Shi. "Privacy-Aware Cost-Effective Scheduling Considering Non-Schedulable Appliances in Smart Home." In 2019 IEEE International Conference on Embedded Software and Systems (ICESS), pp. 1-8. IEEE, 2019.
Li, Boyang, Changhao Chenli, Xiaowei Xu, Taeho Jung, and Yiyu Shi. "Exploiting computation power of blockchain for biomedical image segmentation." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 0-0. 2019.
Gong, Yuan, Boyang Li, Christian Poellabauer, and Yiyu Shi. "Real-time adversarial attacks." In Proceedings of the 28th International Joint Conference on Artificial Intelligence, pp. 4672-4680. AAAI Press, 2019.
Boyang Li, Changhao Chenli, Xiaowei Xu, Yiyu Shi and Taeho Jung, "DLBC: A Deep Learning-Based Consensus in Blockchains for Deep Learning Services," IEEE Transactions on Service Computing (under review)
NSF IUCRC at 701 W Main St, Durham, NC 27701, Durham, NC
Conference at Las Vegas Convention Center, Nevada, United States, Las Vegas, Nevada
Conference at Long Beach Convention Center in Long Beach, CA, Long Beach, CA
NSF IUCRC at Salon C/Smith Ballroom, Morris Inn, Notre Dame, IN 46556, USA, Notre Dame, IN