1 code implementation • 9 Sep 2024 • Qiang Huang, Xiao Yan, Xin Wang, Susie Xi Rao, Zhichao Han, Fangcheng Fu, Wentao Zhang, Jiawei Jiang
We also adapt Transformer codebase to train TF-TGN efficiently with multiple GPUs.
no code implementations • 1 Sep 2024 • Yuxiang Wang, Xiao Yan, Shiyu Jin, Quanqing Xu, Chuanhui Yang, Yuanyuan Zhu, Chuang Hu, Bo Du, Jiawei Jiang
Text matching retrieves texts with similar embeddings to match with a node.
no code implementations • 3 Aug 2024 • Qinbo Zhang, Xiao Yan, Yukai Ding, Quanqing Xu, Chuang Hu, Xiaokai Zhou, Jiawei Jiang
As such, we propose TreeCSS as an efficient VFL framework that accelerates the two main steps.
no code implementations • 8 May 2024 • Renjie Liu, Yichuan Wang, Xiao Yan, Zhenkun Cai, Minjie Wang, Haitian Jiang, Bo Tang, Jinyang Li
In particular, by conducting graph sampling beforehand, DiskGNN acquires the node features that will be accessed by model computation, and such information is utilized to pack the target node features contiguously on disk to avoid read amplification.
1 code implementation • 12 Feb 2024 • Wentao Ning, Reynold Cheng, Xiao Yan, Ben Kao, Nan Huo, Nur AI Hasan Haldar, Bo Tang
Many methods have been proposed to reduce GP bias but they fail to notice the fundamental problem of GP, i. e., it considers popularity from a \textit{global} perspective of \textit{all users} and uses a single set of popular items, and thus cannot capture the interests of individual users.
no code implementations • 4 Jan 2024 • Jingying Zeng, Richard Huang, Waleed Malik, Langxuan Yin, Bojan Babic, Danny Shacham, Xiao Yan, Jaewon Yang, Qi He
First is knowledge tasks where users want to find new knowledge and information, such as search and question-answering.
1 code implementation • 16 Dec 2023 • Yuntao Gui, Xiao Yan, Peiqi Yin, Han Yang, James Cheng
Thus, we design the sparse MHA module, which computes and stores only large attention weights to reduce memory consumption, and the routed FFN module, which dynamically activates a subset of model parameters for each token to reduce computation cost.
no code implementations • 16 Dec 2023 • Jingying Zeng, Jaewon Yang, Waleed Malik, Xiao Yan, Richard Huang, Qi He
Second, there is a concern with the quality of the content generative AI produces, which often lacks the distinctiveness and authenticity that human-created content possesses.
no code implementations • 24 Oct 2023 • Yuxiang Wang, Xiao Yan, Chuang Hu, Fangcheng Fu, Wentao Zhang, Hao Wang, Shuo Shang, Jiawei Jiang
For graph self-supervised learning (GSSL), masked autoencoder (MAE) follows the generative paradigm and learns to reconstruct masked graph edges or node features.
no code implementations • 19 Oct 2023 • Haitian Jiang, Renjie Liu, Xiao Yan, Zhenkun Cai, Minjie Wang, David Wipf
Among the many variants of graph neural network (GNN) architectures capable of modeling data with cross-instance relations, an important subclass involves layers designed such that the forward pass iteratively reduces a graph-regularized energy function of interest.
1 code implementation • 10 Aug 2023 • Wentao Ning, Xiao Yan, Weiwen Liu, Reynold Cheng, Rui Zhang, Bo Tang
We propose a new MDR method named EDDA with two key components, i. e., embedding disentangling recommender and domain alignment, to tackle the two challenges respectively.
no code implementations • 28 Nov 2022 • Peiqi Yin, Xiao Yan, Jinjing Zhou, Qiang Fu, Zhenkun Cai, James Cheng, Bo Tang, Minjie Wang
In this paper, we develop Deep Graph Inference (DGI) -- a system for easy and efficient GNN model inference, which automatically translates the training code of a GNN model for layer-wise execution.
1 code implementation • CVPR 2022 • Dan Zeng, Zhiyuan Lin, Xiao Yan, YuTing Liu, Fei Wang, Bo Tang
To combat the mismatch between FR and FER data, Meta-Face2Exp uses a circuit feedback mechanism, which improves the base network with the feedback from the adaptation network.
no code implementations • 23 Dec 2021 • Wentao Ning, Reynold Cheng, Jiajun Shen, Nur Al Hasan Haldar, Ben Kao, Xiao Yan, Nan Huo, Wai Kit Lam, Tian Li, Bo Tang
Specifically, we define a vector encoding for meta-paths and design a policy network to extend meta-paths.
no code implementations • 3 Nov 2021 • Xiao Yan, Xianghua Gan, Jingjing Tang, Rui Wang
When pre-trained on the big scale datasets and transferred to the medium and small scale datasets, ProSTformer achieves a significant enhancement and behaves best.
no code implementations • 3 Aug 2021 • Bowen Huang, Xiao Yan, Jinjia Zhou, Yibo Fan
Most deep network methods for compressive sensing reconstruction suffer from the black-box characteristic of DNN.
no code implementations • Proceedings of the 2021 International Conference on Management of Data 2021 • Yidi Wu, Yuntao Gui, Tatiana Jin, James Cheng, Xiao Yan, Peiqi Yin, Yufei Cai, Bo Tang, Fan Yu
Graph neural networks (GNNs) have achieved remarkable performance in many graph analytics tasks such as node classification, link prediction and graph clustering.
1 code implementation • 19 May 2021 • Jiansheng Fang, Huazhu Fu, Dan Zeng, Xiao Yan, Yuguang Yan, Jiang Liu
When encountering a dubious diagnostic case, medical instance retrieval can help radiologists make evidence-based diagnoses by finding images containing instances similar to a query case from a large image database.
1 code implementation • Proceedings of the Sixteenth European Conference on Computer Systems 2021 • Zhenkun Cai, Xiao Yan, Yidi Wu, Kaihao Ma, James Cheng, Fan Yu
Graph neural networks (GNNs) have gained increasing popularity in many areas such as e-commerce, social networks and bio-informatics.
no code implementations • IEEE Transactions on Parallel and Distributed Systems 2021 • Yidi Wu, Kaihao Ma, Xiao Yan, Zhi Liu, Zhenkun Cai, Yuzhen Huang, James Cheng, Han Yuan, Fan Yu
We study how to support elasticity, that is, the ability to dynamically adjust the parallelism (i. e., the number of GPUs), for deep neural network (DNN) training in a GPU cluster.
no code implementations • 1 Jan 2021 • An Xu, Xiao Yan, Hongchang Gao, Heng Huang
The heavy communication for model synchronization is a major bottleneck for scaling up the distributed deep neural network training to many workers.
no code implementations • 27 Oct 2020 • Yitong Meng, Jie Liu, Xiao Yan, James Cheng
When a new user just signs up on a website, we usually have no information about him/her, i. e. no interaction with items, no user profile and no social links with other users.
no code implementations • 23 Sep 2020 • Xiao Yan, Gang Kou, Feng Xiao, Dapeng Zhang, Xianghua Gan
Spatial and temporal features are critical for demand forecasting in BSSs, but it is challenging to extract spatiotemporal dynamics.
1 code implementation • 16 Apr 2020 • Zhenkun Cai, Kaihao Ma, Xiao Yan, Yidi Wu, Yuzhen Huang, James Cheng, Teng Su, Fan Yu
A good parallelization strategy can significantly improve the efficiency or reduce the cost for the distributed training of deep neural networks (DNNs).
1 code implementation • 18 Feb 2020 • Han Yang, Xiao Yan, Xinyan Dai, Yongqiang Chen, James Cheng
In this paper, we propose self-enhanced GNN (SEG), which improves the quality of the input data using the outputs of existing GNN models for better performance on semi-supervised node classification.
2 code implementations • 31 Jan 2020 • Xinyan Dai, Xiao Yan, Kaiwen Zhou, Yuxuan Wang, Han Yang, James Cheng
Edit-distance-based string similarity search has many applications such as spell correction, data de-duplication, and sequence alignment.
1 code implementation • 12 Nov 2019 • Xinyan Dai, Xiao Yan, Kaiwen Zhou, Han Yang, Kelvin K. W. Ng, James Cheng, Yu Fan
In particular, at the high compression ratio end, HSQ provides a low per-iteration communication cost of $O(\log d)$, which is favorable for federated learning.
2 code implementations • 12 Nov 2019 • Xinyan Dai, Xiao Yan, Kelvin K. W. Ng, Jie Liu, James Cheng
In this paper, we present a new angle to analyze the quantization error, which decomposes the quantization error into norm error and direction error.
no code implementations • 30 Sep 2019 • Jie Liu, Xiao Yan, Xinyan Dai, Zhirong Li, James Cheng, Ming-Chang Yang
Then we explain the good performance of ip-NSW as matching the norm bias of the MIPS problem - large norm items have big in-degrees in the ip-NSW proximity graph and a walk on the graph spends the majority of computation on these items, thus effectively avoids unnecessary computation on small norm items.
no code implementations • 10 Sep 2019 • Yitong Meng, Xinyan Dai, Xiao Yan, James Cheng, Weiwen Liu, Benben Liao, Jun Guo, Guangyong Chen
Collaborative filtering, a widely-used recommendation technique, predicts a user's preference by aggregating the ratings from similar users.
1 code implementation • 22 Oct 2018 • Xiao Yan, Xinyan Dai, Jie Liu, Kaiwen Zhou, James Cheng
Recently, locality sensitive hashing (LSH) was shown to be effective for MIPS and several algorithms including $L_2$-ALSH, Sign-ALSH and Simple-LSH have been proposed.
1 code implementation • NeurIPS 2018 • Xiao Yan, Jinfeng Li, Xinyan Dai, Hongzhi Chen, James Cheng
Neyshabur and Srebro proposed Simple-LSH, which is the state-of-the-art hashing method for maximum inner product search (MIPS) with performance guarantee.
no code implementations • 22 Jul 2018 • Xiao Yan, Guo Jiafeng, Fan Yixing, Lan Yanyan, Xu Jun, Cheng Xueqi
Our experiments show that both hybrid index and search schemes can improve the recall of the initial retrieval stage with small overhead.