ICLR 2023 accepted papers on Time series
23 of the 1,575 papers accepted at ICLR 2023 (1.5%) are on Time series, compared with 2.3% at ICLR 2022. They fall into 4 topics.
See ICLR 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Causal and generative time series · 10 papers
- Learning Fast and Slow for Online Time Series Forecasting (15 citations)
- Koopman Neural Operator Forecaster for Time-series with Temporal Distributional Shifts (6 citations)
- Temporal Dependencies in Feature Importance for Time Series Prediction (6 citations)
- Anamnesic Neural Differential Equations with Orthogonal Polynomial Projections (2 citations)
- Robust Multivariate Time-Series Forecasting: Adversarial Attacks and Defense Mechanisms (2 citations)
- Phase2vec: dynamical systems embedding with a physics-informed convolutional network (1 citation)
- Spatio-temporal point processes with deep non-stationary kernels (1 citation)
- Meta Temporal Point Processes
- Recursive Time Series Data Augmentation
- Weighted Clock Logic Point Process
Long-term forecasting · 9 papers
- A Time Series is Worth 64 Words: Long-term Forecasting with Transformers (554 citations)
- TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis (439 citations)
- Scaleformer: Iterative Multi-scale Refining Transformers for Time Series Forecasting (37 citations)
- Effectively Modeling Time Series with Simple Discrete State Spaces (15 citations)
- Liquid Structural State-Space Models (13 citations)
- Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting
- MICN: Multi-scale Local and Global Context Modeling for Long-term Series Forecasting
- Multivariate Time-series Imputation with Disentangled Temporal Representations
- Sequential Latent Variable Models for Few-Shot High-Dimensional Time-Series Forecasting