NeurIPS 2023 accepted papers on Time series
52 of the 3,584 papers accepted at NeurIPS 2023 (1.5%) are on Time series, compared with 0.9% at NeurIPS 2022. They fall into 4 topics.
See NeurIPS 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 · 23 papers
- Koopa: Learning Non-stationary Time Series Dynamics with Koopman Predictors (42 citations)
- Conformal PID Control for Time Series Prediction (27 citations)
- ContiFormer: Continuous-Time Transformer for Irregular Time Series Modeling (27 citations)
- OneNet: Enhancing Time Series Forecasting Models under Concept Drift by Online Ensembling (21 citations)
- Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series Forecasting (18 citations)
- Conformal Prediction for Time Series with Modern Hopfield Networks (11 citations)
- WildfireSpreadTS: A dataset of multi-modal time series for wildfire spread prediction (11 citations)
- On the Constrained Time-Series Generation Problem (7 citations)
- Homotopy-based training of NeuralODEs for accurate dynamics discovery (4 citations)
- Curve Your Enthusiasm: Concurvity Regularization in Differentiable Generalized Additive Models (3 citations)
- Sparse Deep Learning for Time Series Data: Theory and Applications (3 citations)
- Bifurcations and loss jumps in RNN training (2 citations)
and 11 more in the interactive map.
Long-term forecasting · 13 papers
- Frequency-domain MLPs are More Effective Learners in Time Series Forecasting (111 citations)
- FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective (61 citations)
- LargeST: A Benchmark Dataset for Large-Scale Traffic Forecasting (29 citations)
- Adaptive Normalization for Non-stationary Time Series Forecasting: A Temporal Slice Perspective (17 citations)
- Sparse Graph Learning from Spatiotemporal Time Series (11 citations)
- Enhancing Adaptive History Reserving by Spiking Convolutional Block Attention Module in Recurrent Neural Networks (7 citations)
- Taming Local Effects in Graph-based Spatiotemporal Forecasting (6 citations)
- Neural Lad: A Neural Latent Dynamics Framework for Times Series Modeling (4 citations)
- SutraNets: Sub-series Autoregressive Networks for Long-Sequence, Probabilistic Forecasting (2 citations)
- Concentration analysis of multivariate elliptic diffusions (1 citation)
- BasisFormer: Attention-based Time Series Forecasting with Learnable and Interpretable Basis
- CrossGNN: Confronting Noisy Multivariate Time Series Via Cross Interaction Refinement
and 1 more in the interactive map.
Foundation models for time series · 13 papers
- One Fits All: Power General Time Series Analysis by Pretrained LM (156 citations)
- Large Language Models Are Zero-Shot Time Series Forecasters (96 citations)
- SimMTM: A Simple Pre-Training Framework for Masked Time-Series Modeling (40 citations)
- Time Series as Images: Vision Transformer for Irregularly Sampled Time Series (21 citations)
- Contrast Everything: A Hierarchical Contrastive Framework for Medical Time-Series (17 citations)
- ForecastPFN: Synthetically-Trained Zero-Shot Forecasting (14 citations)
- Finding Order in Chaos: A Novel Data Augmentation Method for Time Series in Contrastive Learning (10 citations)
- Improving *day-ahead* Solar Irradiance Time Series Forecasting by Leveraging Spatio-Temporal Context (9 citations)
- FOCAL: Contrastive Learning for Multimodal Time-Series Sensing Signals in Factorized Orthogonal Latent Space (8 citations)
- BioMassters: A Benchmark Dataset for Forest Biomass Estimation using Multi-modal Satellite Time-series (5 citations)
- Synthcity: a benchmark framework for diverse use cases of tabular synthetic data (5 citations)
- Scale-teaching: Robust Multi-scale Training for Time Series Classification with Noisy Labels (4 citations)
and 1 more in the interactive map.
Anomaly detection · 3 papers
- MEMTO: Memory-guided Transformer for Multivariate Time Series Anomaly Detection (29 citations)
- Drift doesn't Matter: Dynamic Decomposition with Diffusion Reconstruction for Unstable Multivariate Time Series Anomaly Detection (17 citations)
- Nominality Score Conditioned Time Series Anomaly Detection by Point/Sequential Reconstruction (11 citations)