# TimesURL: Self-supervised Contrastive Learning for Universal Time Series Representation Learning

arXiv: https://arxiv.org/abs/2312.15709 · HTML: https://mlsanity.com/paper/2312.15709/

## Key facts

- Cited by 52 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 45 references that are in the map.
- First posted on arXiv 25 Dec 2023.

## Builds on (most cited first, 29 of 29)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://mlsanity.com/paper/2012.07436/) (2020)
- [Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting](https://mlsanity.com/paper/2106.13008/) (2021)
- [An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling](https://mlsanity.com/paper/1803.01271/) (2018)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [N-BEATS: Neural basis expansion analysis for interpretable time series forecasting](https://mlsanity.com/paper/1905.10437/) (2019)
- [Contrastive Learning with Hard Negative Samples](https://mlsanity.com/paper/2010.04592/) (2020)
- [Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting](https://mlsanity.com/paper/1907.00235/) (2019)
- [TS2Vec: Towards Universal Representation of Time Series](https://mlsanity.com/paper/2106.10466/) (2021)
- [A Transformer-based Framework for Multivariate Time Series Representation Learning](https://mlsanity.com/paper/2010.02803/) (2020)
- [The UCR Time Series Archive](https://mlsanity.com/paper/1810.07758/) (2018)
- [Hard Negative Mixing for Contrastive Learning](https://mlsanity.com/paper/2010.01028/) (2020)
- [A Recurrent Latent Variable Model for Sequential Data](https://mlsanity.com/paper/1506.02216/) (2015)
- [Time-Series Representation Learning via Temporal and Contextual Contrasting](https://mlsanity.com/paper/2106.14112/) (2021)
- [ROCKET: Exceptionally fast and accurate time series classification using random convolutional kernels](https://mlsanity.com/paper/1910.13051/) (2019)
- [SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interaction](https://mlsanity.com/paper/2106.09305/) (2021)
- [Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications](https://mlsanity.com/paper/1802.03903/) (2018)
- [Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding](https://mlsanity.com/paper/2106.00750/) (2021)
- [Structured Inference Networks for Nonlinear State Space Models](https://mlsanity.com/paper/1609.09869/) (2016)
- [Time-Series Anomaly Detection Service at Microsoft](https://mlsanity.com/paper/1906.03821/) (2019)
- [Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features](https://mlsanity.com/paper/1703.02507/) (2017)
- [Time Series Contrastive Learning with Information-Aware Augmentations](https://mlsanity.com/paper/2303.11911/) (2023)
- [Are all negatives created equal in contrastive instance discrimination?](https://mlsanity.com/paper/2010.06682/) (2020)
- [TimeNet: Pre-trained deep recurrent neural network for time series classification](https://mlsanity.com/paper/1706.08838/) (2017)
- [Negative Sampling for Contrastive Representation Learning: A Review](https://mlsanity.com/paper/2206.00212/) (2022)
- [Mind the Gap when Conditioning Amortised Inference in Sequential Latent-Variable Models](https://mlsanity.com/paper/2101.07046/) (2021)

## Built on by (most cited first, 30 of 52)

- [Universal Time-Series Representation Learning: A Survey](https://mlsanity.com/paper/2401.03717/) (2024)
- [Merlin: Multi-View Representation Learning for Robust Multivariate Time Series Forecasting with Unfixed Missing Rates](https://mlsanity.com/paper/2506.12459/) (2025)
- [Bridging Past and Future: Distribution-Aware Alignment for Time Series Forecasting](https://mlsanity.com/paper/2509.14181/) (2025)
- [PatchAD: A Lightweight Patch-based MLP-Mixer for Time Series Anomaly Detection](https://mlsanity.com/paper/2401.09793/) (2024)
- [Frequency-Masked Embedding Inference: A Non-Contrastive Approach for Time Series Representation Learning](https://mlsanity.com/paper/2412.20790/) (2024)
- [AimTS: Augmented Series and Image Contrastive Learning for Time Series Classification](https://mlsanity.com/paper/2504.09993/) (2025)
- [Revitalizing Canonical Pre-Alignment for Irregular Multivariate Time Series Forecasting](https://mlsanity.com/paper/2508.01971/) (2025)
- [Mask the Redundancy: Evolving Masking Representation Learning for Multivariate Time-Series Clustering](https://mlsanity.com/paper/2511.17008/) (2025)
- [Automated Contrastive Learning Strategy Search for Time Series](https://mlsanity.com/paper/2403.12641/) (2024)
- [UniCL: A Universal Contrastive Learning Framework for Large Time Series Models](https://mlsanity.com/paper/2405.10597/) (2024)
- [TopoCL: Topological Contrastive Learning for Time Series](https://mlsanity.com/paper/2502.02924/) (2025)
- [Not All Data are Good Labels: On the Self-supervised Labeling for Time Series Forecasting](https://mlsanity.com/paper/2502.14704/) (2025)
- [Fuzzy Cluster-Aware Contrastive Clustering for Time Series](https://mlsanity.com/paper/2503.22211/) (2025)
- [How Different from the Past? Spatio-Temporal Time Series Forecasting with Self-Supervised Deviation Learning](https://mlsanity.com/paper/2510.04908/) (2025)
- [Training-Free Time Series Classification via In-Context Reasoning with LLM Agents](https://mlsanity.com/paper/2510.05950/) (2025)
- [The Multiverse of Time Series Machine Learning: an Archive for Multivariate Time Series Classification](https://mlsanity.com/paper/2603.20352/) (2026)
- [One-Embedding-Fits-All: Efficient Zero-Shot Time Series Forecasting by a Model Zoo](https://mlsanity.com/paper/2509.04208/) (2025)
- [Beyond Fixed Variables: Expanding-variate Time Series Forecasting via Flat Scheme and Spatio-temporal Focal Learning](https://mlsanity.com/paper/2502.15296/) (2025)
- [Orthogonalized Multimodal Contrastive Learning with Asymmetric Masking for Structured Representations](https://mlsanity.com/paper/2602.14983/) (2026)
- [HEPA: A Self-Supervised Horizon-Conditioned Event Predictive Architecture for Time Series](https://mlsanity.com/paper/2605.11130/) (2026)
- [Content-aware Balanced Spectrum Encoding in Masked Modeling for Time Series Classification](https://mlsanity.com/paper/2412.13232/) (2024)
- [Structure-preserving contrastive learning for spatial time series](https://mlsanity.com/paper/2502.06380/) (2025)
- [Dual-Splitting Conformal Prediction for Multi-Step Time Series Forecasting](https://mlsanity.com/paper/2503.21251/) (2025)
- [Enhancing Contrastive Learning-based Electrocardiogram Pretrained Model with Patient Memory Queue](https://mlsanity.com/paper/2506.06310/) (2025)
- [NeuCoReClass AD: Redefining Self-Supervised Time Series Anomaly Detection](https://mlsanity.com/paper/2508.00909/) (2025)
- [Instruction-based Time Series Editing](https://mlsanity.com/paper/2508.01504/) (2025)
- [Learning Time-Series Representations by Hierarchical Uniformity-Tolerance Latent Balancing](https://mlsanity.com/paper/2510.01658/) (2025)
- [CF-JEPA: Mask-free forward prediction with asymmetric encoder utilization for time-series representation learning](https://mlsanity.com/paper/2606.07031/) (2026)
- [Efficient Time Series Clustering from Multiscale Reservoir Dynamics with Granular-Ball Anchoring Graph Optimization](https://mlsanity.com/paper/2606.12077/) (2026)
- [LLM Agents for Time-Series: A Survey](https://mlsanity.com/paper/2608.26226/) (2026)

## Abstract (arXiv)

Learning universal time series representations applicable to various types of downstream tasks is challenging but valuable in real applications. Recently, researchers have attempted to leverage the success of self-supervised contrastive learning (SSCL) in Computer Vision(CV) and Natural Language Processing(NLP) to tackle time series representation. Nevertheless, due to the special temporal characteristics, relying solely on empirical guidance from other domains may be ineffective for time series and difficult to adapt to multiple downstream tasks. To this end, we review three parts involved in SSCL including 1) designing augmentation methods for positive pairs, 2) constructing (hard) negative pairs, and 3) designing SSCL loss. For 1) and 2), we find that unsuitable positive and negative pair construction may introduce inappropriate inductive biases, which neither preserve temporal properties nor provide sufficient discriminative features. For 3), just exploring segment- or instance-level semantics information is not enough for learning universal representation. To remedy the above issues, we propose a novel self-supervised framework named TimesURL. Specifically, we first introduce a frequency-temporal-based augmentation to keep the temporal property unchanged. And then, we construct double Universums as a special kind of hard negative to guide better contrastive learning. Additionally, we introduce time reconstruction as a joint optimization objective with contrastive learning to capture both segment-level and instance-level information. As a result, TimesURL can learn high-quality universal representations and achieve state-of-the-art performance in 6 different downstream tasks, including short- and long-term forecasting, imputation, classification, anomaly detection and transfer learning.