When the attention matrix is obtained, it is formulated into a propagation matrix for graph neural networks. Finally, a Time-extrapolator Convolutional Neural Network (TXP-CNN) is used in the temporal dimension of the aggregated features to predict the future trajectories of the pedestrians. ...
这篇文章是华为诺亚实验室在GCN的基础上,加入邻居节点交互感知完成的一个推荐系统方向的论文,提出了模型NIA-GCN,被发表在SIGIR 2020上,文中关于邻居交互感知的部分,与BGNN模型比较像(可以参考本人CSDN《论文阅读》专栏 - 『论文《Bilinear Graph Neural Network with Neighbor Interactions》阅读』一文)。 论文地址:...
When the attention matrix is obtained, it is formulated into a propagation matrix for graph neural networks. Finally, a Time-extrapolator Convolutional Neural Network (TXP-CNN) is used in the temporal dimension of the aggregated features to predict the future trajectories of the pedestrians. ...
GraphReg (Chromatin interaction aware gene regulatory modeling with graph attention networks) is a graph neural network based gene regulation model which integrates DNA sequence, 1D epigenomic data (such as chromatin accessibility and histone modifications), and 3D chromatin conformation data (such as Hi...
Learning Physical Dynamics with Subequivariant Graph Neural Networks, NeurIPS 2022. [paper] [code] EvolveGraph: Multi-Agent Trajectory Prediction with Dynamic Relational Reasoning, NeurIPS 2020. [paper] Interaction Templates for Multi-Robot Systems, IROS 2019. [paper] Factorised Neural Relational Inferenc...
DialogueGCN: 一种用于会话情感识别的图形卷积神经网络 DialogueGCN: A Graph Convolutional Neural Network for Emotion Recognition in Conversation 语言环境建模的重要性,序列模型现有缺点是还是遗忘问题,从图的方面,改进了序列模型的缺点 0. Abstract 对话情感识别(Emotion recogn...猜...
Learning human-object interactions by graph parsing neural networks. In Proceedings of the Eu- ropean conference on computer vision (ECCV), pages 401– 417, 2018. 2 [30] Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Giri...
Learning human-object interactions by graph parsing neural networks. In ECCV, 2018. 1, 2, 6, 7 [33] Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster r-cnn: Towards real-time object detection with region proposal networks. N...
(RL) based methods fail to consider different types of interactions among all agents or ignore the intentions of people, which results in performance degradation. In this paper, we propose a novel recurrent graph neural network with attention mechanisms to capture heterogeneous interactions among ...
Neural Relational Inference for Interacting Systems, ICML 2018. [paper] [code] Unsupervised Learning of Latent Physical Properties Using Perception-Prediction Networks, UAI 2018. [paper] Relational inductive biases, deep learning, and graph networks, 2018. [paper] Relational Neural Expectation Maximization...