2021年三大顶会时间序列预测论文&代码整理

转载：炼丹笔记

2021年最新时间序列预测论文&代码整理

AAAI 2021

1. Deep Switching Auto-Regressive Factorization: Application to Time Series Forecasting
   • 下载：https://arxiv.org/abs/2009.05135
   • 代码：https://github.com/ostadabbas/DSARF
2. Dynamic Gaussian Mixture Based Deep Generative Model for Robust Forecasting on Sparse Multivariate Time Series
   • 下载：https://arxiv.org/abs/2103.02164
   • 代码：https://paperswithcode.com/paper/dynamic-gaussian-mixture-based-deep#code
3. Temporal Latent Autoencoder: A Method for Probabilistic Multivariate Time Series Forecasting
   • 下载：https://arxiv.org/abs/2101.10460
   • 代码：未找到
4. Synergetic Learning of Heterogeneous Temporal Sequences for Multi-Horizon Probabilistic Forecasting
   • 下载：https://arxiv.org/abs/2102.00431
   • 代码：未找到
5. Correlative Channel-Aware Fusion for Multi-View Time Series Classification
   • 下载：https://arxiv.org/abs/1911.11561
   • 代码：未找到
6. Learnable Dynamic Temporal Pooling for Time Series Classification
   • 下载：https://arxiv.org/abs/2104.02577
   • 代码：https://github.com/donalee/DTW-Pool
7. ShapeNet: A Shapelet-Neural Network Approach for Multivariate Time Series Classification
   • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/17018
   • 代码：未找到
8. Joint-Label Learning by Dual Augmentation for Time Series Classification
   • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/17071
   • 代码：未找到
9. Graph Neural Network-Based Anomaly Detection in Multivariate Time Series
   • 下载：https://arxiv.org/abs/2106.06947
   • 代码：https://github.com/d-ailin/GDN
10. Time Series Anomaly Detection with Multiresolution Ensemble Decoding
    • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/17152
    • 代码：未找到
11. Outlier Impact Characterization for Time Series Data
    • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/17379
    • 代码：未找到
12. Generative Semi-Supervised Learning for Multivariate Time Series Imputation
    • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/17086
    • 代码：https://githubmemory.com/repo/zjuwuyy-DL/Generative-Semi-supervised-Learning-for-Multivariate-Time-Series-Imputation
13. Bridging Towers of Multi-Task Learning with a Gating Mechanism for Aspect-Based Sentiment Analysis and Sequential Metaphor Identification
    • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/17596
    • 代码：未找到
14. C2F-FWN: Coarse-to-Fine Flow Warping Network for Spatial-Temporal Consistent Motion Transfer
    • 下载：https://arxiv.org/abs/2012.08976
    • 代码：https://github.com/wswdx/C2F-FWN
15. Inductive Graph Neural Networks for Spatiotemporal Kriging
    • 下载：https://arxiv.org/abs/2006.07527
    • 代码：https://github.com/Kaimaoge/IGNNK
16. Temporal-Coded Deep Spiking Neural Network with Easy Training and Robust Performance
    • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/17329
    • 代码：https://github.com/zbs881314/Temporal-Coded-Deep-SNN
17. Continuous-Time Attention for Sequential Learning
    • 下载：https://ojs.aaai.org/index.php/AAAI/article/view/16875
    • 代码：未找到
18. ChronoR: Rotation Based Temporal Knowledge Graph Embedding
    • 下载：https://arxiv.org/abs/2103.10379
    • 代码：未找到
19. Learning from History: Modeling Temporal Knowledge Graphs with Sequential CopyGeneration Networks
    • 下载：https://arxiv.org/abs/2012.08492
    • 代码：未找到
20. Neural Latent Space Model for Dynamic Networks and Temporal Knowledge Graphs
    • 下载：https://arxiv.org/abs/1911.11455
    • 代码：未找到

ICML 2021

1. Voice2Series: Reprogramming Acoustic Models for Time Series Classification
   • 下载：https://arxiv.org/abs/2106.09296
   • 代码：https://github.com/huckiyang/Voice2Series-Reprogramming
2. Neural Rough Differential Equations for Long Time Series
   • 下载：https://arxiv.org/abs/2009.08295
   • 代码：https://github.com/jambo6/neuralRDEs
3. Necessary and sufficient conditions for causal feature selection in time series with latent common causes
   • 下载：https://arxiv.org/abs/2005.08543
   • 代码：未找到
4. Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting
   • 下载：https://arxiv.org/abs/2101.12072
   • 代码：未找到
5. Conformal prediction interval for dynamic time-series
   • 下载：https://arxiv.org/abs/2010.09107
   • 代码：https://github.com/hamrel-cxu/EnbPI
6. Z-GCNETs: Time Zigzags at Graph Convolutional Networks for Time Series Forecasting
   • 下载：https://arxiv.org/abs/2105.04100
   • 代码：https://github.com/Z-GCNETs/Z-GCNETs
7. End-to-End Learning of Coherent Probabilistic Forecasts for Hierarchical Time Series
   • 下载：https://proceedings.mlr.press/v139/rangapuram21a.html
   • 代码：https://github.com/awslabs/gluon-ts
8. Approximation Theory of Convolutional Architectures for Time Series Modelling
   • 下载：http://proceedings.mlr.press/v139/jiang21d/jiang21d.pdf
   • 代码：未找到
9. Whittle Networks: A Deep Likelihood Model for Time Series
   • 下载：http://proceedings.mlr.press/v139/yu21c.html
   • 代码：https://github.com/ml-research/WhittleNetworks
10. Explaining Time Series Predictions with Dynamic Masks
    • 下载：https://arxiv.org/abs/2106.05303
    • 代码：https://github.com/JonathanCrabbe/Dynamask
11. ST-DETR: Spatio-Temporal Object Traces Attention Detection Transformer

- 下载：https://arxiv.org/pdf/2107.05887.pdf

- 代码：未找到

1. Temporal Dependencies in Feature Importance for Time Series Predictions
   • 下载：https://arxiv.org/abs/2107.14317
   • 代码：未找到

IJCAI

1. Time-Aware Multi-Scale RNNs for Time Series Modeling
   • 下载：https://www.ijcai.org/proceedings/2021/315
   • 代码：https://github.com/qianlima-lab/TAMS-RNNs
2. Two Birds with One Stone: Series Saliency for Accurate and Interpretable Multivariate Time Series Forecasting
   • 下载：https://www.ijcai.org/proceedings/2021/397
   • 代码：未找到
3. TE-ESN: Time Encoding Echo State Network for Prediction Based on Irregularly Sampled Time Series Data
   • 下载：https://arxiv.org/abs/2105.00412
   • 代码：未找到
4. Time-Series Representation Learning via Temporal and Contextual Contrasting
   • 下载：https://arxiv.org/abs/2106.14112
   • 代码：https://github.com/emadeldeen24/TS-TCC
5. Time Series Data Augmentation for Deep Learning: A Survey
   • 下载：https://arxiv.org/abs/2002.12478
   • 代码：无
6. Uncertain Time Series Classification
   • 下载：https://www.ijcai.org/proceedings/2021/0683.pdf
   • 代码：https://github.com/frankl1/ustc
7. Learning Temporal Causal Sequence Relationships from Real-Time Time-
   • 下载：https://arxiv.org/abs/1905.12262
   • 代码：未找到
8. Adversarial Spectral Kernel Matching for Unsupervised Time Series Domain Adaptation
   • 下载：https://www.ijcai.org/proceedings/2021/378
   • 代码：https://github.com/jarheadjoe/Adv-spec-ker-matching
9. Multi-series Time-aware Sequence Partitioning for Disease Progression Modeling
   • 下载：https://www.ijcai.org/proceedings/2021/493
   • 代码：未找到

参考文献

1. https://www.yanxishe.com/reportDetail/26029
2. https://icml.cc/Conferences/2021/Schedule?type=Poster
3. https://ijcai-21.org/program-main-track/
4. https://dreamhomes.top/posts/202108241839