Pith. sign in

REVIEW 10 cited by

Time-Series Representation Learning via Temporal and Contextual Contrasting

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2106.14112 v1 pith:OU5LQET3 submitted 2021-06-26 cs.LG cs.AI

Emadeldeen Eldele , Mohamed Ragab , Zhenghua Chen , Min Wu , Chee Keong Kwoh , Xiaoli Li , Cuntai Guan This is my paper
classification cs.LGcs.AI
keywords temporaltime-seriescontrastingdatalearningts-tcccontextscontextual
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Learning decent representations from unlabeled time-series data with temporal dynamics is a very challenging task. In this paper, we propose an unsupervised Time-Series representation learning framework via Temporal and Contextual Contrasting (TS-TCC), to learn time-series representation from unlabeled data. First, the raw time-series data are transformed into two different yet correlated views by using weak and strong augmentations. Second, we propose a novel temporal contrasting module to learn robust temporal representations by designing a tough cross-view prediction task. Last, to further learn discriminative representations, we propose a contextual contrasting module built upon the contexts from the temporal contrasting module. It attempts to maximize the similarity among different contexts of the same sample while minimizing similarity among contexts of different samples. Experiments have been carried out on three real-world time-series datasets. The results manifest that training a linear classifier on top of the features learned by our proposed TS-TCC performs comparably with the supervised training. Additionally, our proposed TS-TCC shows high efficiency in few-labeled data and transfer learning scenarios. The code is publicly available at https://github.com/emadeldeen24/TS-TCC.

Discussion (0). Sign in to comment.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Masked Generative-Contrastive Representation Learning for Cross-Dataset EEG-Based Emotion Recognition

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Pretraining a region-aware spatiotemporal EEG encoder with JEPA generative and masked dynamic contrastive losses on FACED yields higher cross-subject accuracy than SSL baselines when fine-tuned on SEED-IV/V/VII.

  2. From Token to Rhythm: A Multi-Scale Approach for ECG-Language Pretraining

    eess.SP 2025-06 conditional novelty 6.0 of 10

    MELP pretrains ECG and text encoders with token-, beat-, and rhythm-level cross-modal supervision and beats prior baselines on several ECG classification benchmarks.

  3. Modular Foundation Models for Time-Series Perception in Digital Twins

    cs.LG 2026-07 conditional novelty 5.0 of 10

    A gated bank of frozen self-supervised time-series encoders, aligned and aggregated by a Transformer, supports competitive multi-task perception for digital twins and hydro-generator virtual sensing.

  4. A Contrastive Diffusion-based Network (CDNet) for Time Series Classification

    cs.LG 2025-07 reject novelty 5.0 of 10

    CDNet generates contrastive training pairs via CNN-based diffusion transitions between time series samples and claims consistent accuracy gains for deep classifiers on binary UCR datasets.

  5. DisMS-TS: Eliminating Redundant Multi-Scale Features for Time Series Classification

    cs.AI 2025-07 conditional novelty 5.0 of 10

    A multi-scale time series classifier that disentangles scale-shared and scale-specific features and reports improved accuracy on six benchmarks.

  6. Transformer-based EEG Decoding: A Survey

    cs.LG 2025-07 conditional novelty 5.0 of 10

    A survey that classifies Transformer-based EEG decoding models into backbone, hybrid, and customized categories and reviews their applications and limitations.

  7. PLanTS: Periodicity-aware Latent-state Representation Learning for Multivariate Time Series

    cs.LG 2025-09 reject novelty 4.0 of 10

    PLanTS combines FFT-driven multi-granularity patching, MXCorr-based soft contrastive loss, and next-transition prediction, claiming state-of-the-art SSL results on UEA, PTB-XL, ETT, and Yahoo benchmarks.

  8. Discrepancy-Aware Contrastive Adaptation in Medical Time Series Analysis

    cs.HC 2025-08 reject novelty 4.0 of 10

    A three-stage pipeline that appends an autoencoder reconstruction error (trained on external normal recordings) to the input and trains attention-based multi-view contrastive representations, claiming SOTA on three EE...

  9. MoSSDA: A Semi-Supervised Domain Adaptation Framework for Multivariate Time-Series Classification using Momentum Encoder

    cs.LG 2025-08 conditional novelty 4.0 of 10

    MoSSDA is a two-stage framework that uses MMD, mixup-based supervised contrastive learning with a momentum encoder, and a frozen-features classifier to improve semi-supervised domain adaptation for time-series classification.

  10. eMargin: Revisiting Contrastive Learning with Margin-Based Separation

    cs.LG 2025-07 reject novelty 4.0 of 10

    An adaptive margin added to InfoNCE improves time series clustering metrics but hurts linear-probe classification, exposing a disconnect between clustering scores and downstream utility.

Pith tools