Haodi Wang
Research Area: Zero-knowledge Proof, Privacy Preservation
TEL:
(+86) 15710068793
February. 1996
E-mail:
lirawang@foxmail.com
Date of Birth:
Education Background
Beijing Normal University
09/2020-Present
09/2018-08/2020
•
•
First year of PhD candidate, recommended;
Research Area: Zero-knowledge proofs
Beijing Normal University
•
•
•
•
Master of Computer Science, recommended;
Research Area: Deep learning and computer vision;
GPA: 3.96 out of 4.0.
Rank: 1 out of 30
Beijing Normal University
09/2014-08/2018
03/2020-01/2021
•
•
•
Bachelor of Science in Computer Science, Department of Artificial Intelligence;
GPA: 3.85 out of 5.0
Rank: 6 out of 54
Research Projects
Research on the proof of deep learning model by zero-knowledge proof
•
•
•
Construct and train the mnist-demo/LeNet-5/AlexNet/VGG16 model;
Give detailed principle and operation of CNN model
Cooperate with SECBIT Labs for the paper (under review).
Research on the batch Plonk zero-knowledge proof scheme
11/2020-Present
12/2020-01/2021
12/2020-Present
09/2018-12/2019
09/2018-08/2019
•
•
Construct the main idea of this project
Draft the paper (revising)
Research on the Halo 0.9 zero-knowledge proof protocol
•
•
Construct the main idea
Report this work on Cryptology ePrint Archive: https://eprint.iacr.org/2020/1573
Research on the privacy-preserving deep learning model
Construct the main idea of this project, it’s an ongoing research and still needs to be revised.
•
Research on image inpainting with generative adversarial network
•
•
Using GAN network and multi-discriminators for image inpainting;
Publish a conference paper and attend to the conference in April, 2020.
Research on image inpainting with text recovery and residual learning
Publish journal papers as co-authors.
•
Research Achievements
Preprint
Cryptology ePrint Archive, Report
2020/1573,
Halo 0.9: A halo protocol with fully-succinctness
https://eprint.iacr.org/2020/1573 (2020)
Semantic Inpainting with Multi-Dimensional Adversarial
Network and Wasserstein Distance.
Conference
paper
3rd Chinese Conference on Pattern
Recognition and Computer Vision
Multi-scale semantic image inpainting with residual learning
and GAN.
journal
paper
Neurocomputing, 2019,
331(9252312):199-212
Text recovery via deep CNN-BiLSTM recognition and
Bayesian inference.
journal paper
IEEE Access, 2018,
6(21693536):76416-76428