# Frequency-Masked Embedding Inference: A Non-Contrastive Approach for Time Series Representation Learning

arXiv: https://arxiv.org/abs/2412.20790 · HTML: https://mlsanity.com/paper/2412.20790/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 27 references that are in the map.
- First posted on arXiv 30 Dec 2024.

## Builds on (most cited first, 15 of 15)

- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [TS2Vec: Towards Universal Representation of Time Series](https://mlsanity.com/paper/2106.10466/) (2021)
- [ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfer](https://mlsanity.com/paper/2105.11741/) (2021)
- [Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding](https://mlsanity.com/paper/2106.00750/) (2021)
- [CERT: Contrastive Self-supervised Learning for Language Understanding](https://mlsanity.com/paper/2005.12766/) (2020)
- [MST: Masked Self-Supervised Transformer for Visual Representation](https://mlsanity.com/paper/2106.05656/) (2021)
- [Relative Flatness and Generalization](https://mlsanity.com/paper/2001.00939/) (2020)
- [Contrast Everything: A Hierarchical Contrastive Framework for Medical Time-Series](https://mlsanity.com/paper/2310.14017/) (2023)
- [TimesURL: Self-supervised Contrastive Learning for Universal Time Series Representation Learning](https://mlsanity.com/paper/2312.15709/) (2023)
- [Contrastive Learning for Unsupervised Domain Adaptation of Time Series](https://mlsanity.com/paper/2206.06243/) (2022)
- [Universal Time-Series Representation Learning: A Survey](https://mlsanity.com/paper/2401.03717/) (2024)
- [Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud](https://mlsanity.com/paper/2404.16432/) (2024)
- [REBAR: Retrieval-Based Reconstruction for Time-series Contrastive Learning](https://mlsanity.com/paper/2311.00519/) (2023)
- [Towards Enhancing Time Series Contrastive Learning: A Dynamic Bad Pair Mining Approach](https://mlsanity.com/paper/2302.03357/) (2023)
- [DMT-JEPA: Discriminative Masked Targets for Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2405.17995/) (2024)

## Built on by (most cited first, 9 of 9)

- [Mask the Redundancy: Evolving Masking Representation Learning for Multivariate Time-Series Clustering](https://mlsanity.com/paper/2511.17008/) (2025)
- [Wavelet-Driven Masked Multiscale Reconstruction for PPG Foundation Models](https://mlsanity.com/paper/2601.12215/) (2026)
- [Learning Clinical Representations Under Systematic Distribution Shift](https://mlsanity.com/paper/2603.07348/) (2026)
- [Discriminative Representation Learning for Clinical Prediction](https://mlsanity.com/paper/2603.20921/) (2026)
- [CF-JEPA: Mask-free forward prediction with asymmetric encoder utilization for time-series representation learning](https://mlsanity.com/paper/2606.07031/) (2026)
- [PEFT-MuTS: A Multivariate Parameter-Efficient Fine-Tuning Framework for Remaining Useful Life Prediction based on Cross-domain Time Series Representation Model](https://mlsanity.com/paper/2601.22631/) (2026)
- [Event Fields: Learning Latent Event Structure for Waveform Foundation Models](https://mlsanity.com/paper/2605.08685/) (2026)
- [Frequency-Domain Multi-Modality Transportation Modeling](https://mlsanity.com/paper/2607.08475/) (2026)
- [Seeing Time: Visual-Temporal Representation Learning for Interpretable Time Series Clustering](https://mlsanity.com/paper/2609.36873/) (2026)

## Abstract (arXiv)

Contrastive learning underpins most current self-supervised time series representation methods. The strategy for constructing positive and negative sample pairs significantly affects the final representation quality. However, due to the continuous nature of time series semantics, the modeling approach of contrastive learning struggles to accommodate the characteristics of time series data. This results in issues such as difficulties in constructing hard negative samples and the potential introduction of inappropriate biases during positive sample construction. Although some recent works have developed several scientific strategies for constructing positive and negative sample pairs with improved effectiveness, they remain constrained by the contrastive learning framework. To fundamentally overcome the limitations of contrastive learning, this paper introduces Frequency-masked Embedding Inference (FEI), a novel non-contrastive method that completely eliminates the need for positive and negative samples. The proposed FEI constructs 2 inference branches based on a prompting strategy: 1) Using frequency masking as prompts to infer the embedding representation of the target series with missing frequency bands in the embedding space, and 2) Using the target series as prompts to infer its frequency masking embedding. In this way, FEI enables continuous semantic relationship modeling for time series. Experiments on 8 widely used time series datasets for classification and regression tasks, using linear evaluation and end-to-end fine-tuning, show that FEI significantly outperforms existing contrastive-based methods in terms of generalization. This study provides new insights into self-supervised representation learning for time series. The code is available at https://github.com/USTBInnovationPark/Frequency-masked-Embedding-Inference.