A Framework for Recommending Accurate and Diverse Items Using Bayesian Graph Convolutional Neural Networks
We take a first step to introduce a principled way to model the uncertainty in the user-item interaction graph using the Bayesian Graph Convolutional Neural Networks framework. We demonstrate that our proposed solution can achieve accurate and diverse recommendation results and alleviate the data sparsity problem when the users have fewer historical interaction records.
Jianing Sun*, et. al. Accepted by ACM SIGKDD, 2020.
Neighbor Interaction Aware Graph Convolution Networks for Recommendation
We propose NIA-GCN, which can explicitly model the relational information between neighbor nodes and exploit the heterogeneous nature of the user-item bipartite graph. Furthermore, we generalize our framework to a commercial App store recommendation scenario. We observe significant improvement on a large-scale commercial dataset.
Jianing Sun*, Yingxue Zhang, Wei Guo, Huifeng Guo, Ruiming Tang, Xiuqiang He, Chen Ma, Mark Coates. Accepted by ACM SIGIR, 2020.
Memory Augmented Graph Neural Networks for Sequential Recommendation
We propose a memory augmented graph neural network (MA-GNN) to capture both the long- and short-term user interests. Specifically, we apply a graph neural network to model the item contextual information within a short-term period and utilize a shared memory network to capture the long-range dependencies between items.
Chen Ma, Liheng Ma, Yingxue Zhang, Jianing Sun*, Xue Liu, Mark Coates. Accepted by AAAI, 2020.
Multi-Graph Convolutional Neural Networks for Representation Learning in Recommendation
We propose a novel graph convolutional neural network (GCNN)-based recommender system framework.
Jianing Sun*, Yingxue Zhang. Accepted by Graph Representation Learning Workshop, NeurIPS, 2019.
Multi-Graph Convolution Collaborative Filtering
We propose a novel collaborative filtering procedure that incorporates multiple graphs to explicitly represent user-item, user-user and item-item relationships. [Slides]
Jianing Sun*, Yingxue Zhang, Chen Ma, Mark Coates, Huifeng Guo, Ruiming Tang, Xiuqiang He. Accepted by IEEE ICDM, 2019.
FoodTracker - A Real-time Mobile App by Deep Convolutional Neural Networks
We develop an automatic mobile app that can regonize food items of multi-object meal from a single image in real-time, and return the nutrition facts with components and approximate amount.
Jianing Sun*, Katarzyna Radecka, Zeljko Zilic. Accepted by The 16th International Conference on Machine Vision Applications, 2019.
Exploring Better Food Detection by Transfer Learning
In this work, we present a deep learning based, food-specialized detection framework with transferring features. [Code]
Jianing Sun*, Katarzyna Radecka, Zeljko Zilic. Accepted by The 16th International Conference on Machine Vision Applications, 2019.
Master Thesis
Master of Engineering, Department of Electrical and Computer Engineering, Jianing Sun, McGill University, May 2019. [Code1] [Code2]