轨迹大模型:轨迹数据表征学习 | 周三直播·时序时空大模型读书会
导语
研究领域:轨迹数据、表征学习、基座模型、深度学习、TrajCL、JEPA、TrajDPM
分享内容简介
分享内容简介
分享内容大纲
分享内容大纲
(一)常晏川:基于深度学习方法的轨迹相似性学习
-
Part 1:问题定义及研究动机
-
Part 2:基于对比学习的深度学习方法 – TrajCL
-
Part 3:方向思考和展望
(二)李立桓:探索JEPA在轨迹表征中的应用
-
Part 1: 基于自监督学习的轨迹相似度匹配
-
Part 2: 联合嵌入预测架构(JEPA)介绍
-
Part 3: 用于轨迹相似度计算的联合嵌入预测架构 – T-JEPA
(三)陈文杰:轨迹表征学习与非参聚类
-
Part1: 轨迹非参聚类背景概览
-
Part2: 轨迹非参聚类模型 – TrajDPM
核心概念
核心概念
-
基座模型 Foundation Model
-
深度学习 Deep Learning
-
轨迹相似性 Trajectory Similarity
-
轨迹查询 Trajectory query
-
轨迹相似度计算 Trajectory Similarity Calculation
-
自监督学习 Self-supervised Learning
-
联合嵌入预测架构 Joint Embedding Prediction Architecture
-
表征学习 Representative Learning
-
非参聚类 Non-parametric Clustering
-
狄利克雷高斯混合模型 Dirichlet Gaussian Mixture Model
分享人介绍
分享人介绍
(1)主讲人:常晏川
(2)主讲人:李立桓
(3)主讲人:陈文杰
(4)主持人:薛昊
本期主要参考文献
本期主要参考文献
-
Li, L., Xue, H., Song, Y. and Salim, F., 2024. T-JEPA: A Joint-Embedding Predictive Architecture for Trajectory Similarity Computation. arXiv preprint arXiv:2406.12913. -
Chang, Y., Qi, J., Liang, Y. and Tanin, E., 2023, April. Contrastive trajectory similarity learning with dual-feature attention. In 2023 IEEE 39th International conference on data engineering (ICDE) (pp. 2933-2945). IEEE. -
Li, X., Zhao, K., Cong, G., Jensen, C.S. and Wei, W., 2018, April. Deep representation learning for trajectory similarity computation. In 2018 IEEE 34th international conference on data engineering (ICDE) (pp. 617-628). IEEE. -
Assran, M., Duval, Q., Misra, I., Bojanowski, P., Vincent, P., Rabbat, M., LeCun, Y. and Ballas, N., 2023. Self-supervised learning from images with a joint-embedding predictive architecture. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 15619-15629). -
Chang, Y., Tanin, E., Cong, G., Jensen, C. S., & Qi, J. (2023). Trajectory Similarity Measurement: An Efficiency Perspective. arXiv preprint arXiv:2311.00960. -
Chen, W., Liang, Y., Zhu, Y., Chang, Y., Luo, K., Wen, H., … & Zheng, Y. (2024). Deep learning for trajectory data management and mining: A survey and beyond. arXiv preprint arXiv:2403.14151. -
LeCun, Y., 2022. A path towards autonomous machine intelligence version 0.9. 2, 2022-06-27. Open Review, 62(1), pp.1-62. -
Yao D., Cong G., Zhang C., et al. Computing Trajectory Similarity in Linear Time: A Generic Seed-Guided Neural Metric Learning Approach. 2019 IEEE 35th International Conference on Data Engineering (ICDE, 2019: 1358–1369) -
Zhang H., Zhang X., Jiang Q., et al. Trajectory similarity learning with auxiliary supervision and optimal matching. https://dl.acm.org/doi/10.5555/3491440.3491884 -
Fang Z., et al. E2DTC:An End to End Deep Trajectory Clustering Framework via Self-Training. https://ieeexplore.ieee.org/document/9458936 -
Chang J., Parallel sampling of DP mixture models using sub-clusters splits. https://dl.acm.org/doi/10.5555/2999611.2999681 -
Meitar Ronen; Shahaf E. Finder; Oren Freifeld. DeepDPM: Deep Clustering With an Unknown Number of Clusters. https://ieeexplore.ieee.org/document/9879746
直播信息
直播信息
时间:
参与方式:
1、集智俱乐部 B 站免费直播,扫码可预约。
2、本期读书会公开进行,扫码即可直接获取读书会回看权限。欢迎感兴趣的朋友报名新一季时序时空大模型读书会,加入群聊,获取系列读书会回看权限,成为“Time-LLMs”社区的种子用户,与社区的一线科研工作者与企业实践者沟通交流,共同推动社区的发展。
报名成为主讲人:
时序时空大模型读书会招募中
点击“阅读原文”,报名读书会