Publications & Projects
Graph-Structured Sequence Modeling through Spatio-Temporal U-Network
Designed a novel multi-scale architecture, Spatio-Temporal U-Net (ST-UNet), for graph-structured time series modeling. In this U-shaped network, a paired sampling operation is proposed in spacetime domain accordingly: the pooling (ST-Pool) and the unpooling (ST-Unpool). To better localize the representation from the input, higher-level features retrieved from the pooling part are concatenated with the upsampled output. The final output of ST-UNet can be utilized for predicting node attributes or the entire graph in the next few time steps.
Yu, B., Yin, H., & Zhu, Z. (2019). ST-UNet: A Spatio-Temporal U-Network for Graph-structured Time Series Modeling. arXiv preprint arXiv:1903.05631. [PDF], * indicates equal contribution.
Machine Learning Attacks to Location Privacy
Developed a model of adversary that uses machine learning to learn about the geographical data collected from users of location-based services and the corresponding privacy mechanism, and then performs its attack on users’ location privacy.
Project finished as a member of the international internship program at École Polytechnique & Inria. This work is currently under preparation for peer review .
Traffic Prediction with Deep Spatial Temporal Neural Nets
Designed a fully integrated convolutional neural network to precisely model the topology of the road network and then accurately forecast the future traffic condition (speed, flow or volume) of the network through space-time series in the mid-and-long term.
Yu, B.*, Yin, H.*, & Zhu, Z. (2018). Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. In Proceedings of the 27th International Joint Conference on Artificial Intelligence (pp. 3634-3640). [Link, PDF (arXiv), Code], * indicates equal contribution.
Neural Artist - Style Transfer in Short Videos
Applied Conditional GANs and Fast Style Transfer to convert short videos into customized styles (e.g. Van Gogh, The Starry Night) through DNN-based texture abstraction and redesigned loss function to balance and minimize the flicker between rendering frames.
This project is awarded 'the Most Technical Difficulty Award' at Schlumberger HackPKU 2017.
Hotspot Prediction Based on Temporal Trajectory and Social Attributes
Proposed a location-vector embedding framework (Loc2vec) to predict the geographic hotspot in a certain area and explore its semantic meaning based on temporal trajectories, which are linked in a chronological order through users’ check-ins gathered from location-based social networks.
- 11/28, '18 - Towards Robust Deep Neural Networks [Slides]
- 04/20, '18 - Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework [Slides]
- 04/06, '18 - Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey [Slides]
- 12/20, '17 - Spatially Transformed Adversarial Examples [Slides]