Research map / Label-efficient and robust learning / Noisy labels
Radar-APLANC: Unsupervised Radar-based Heartbeat Sensing via Augmented Pseudo-Label and Noise Contrast
Ying Wang, Zhaodong Sun, Xu Cheng, Zuxian He, Xiaobai Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add noise matrix as negative sample alongside pseudo-label (stage one) (ablation) | Radar-APLANC · Equipleth | MAE: 8.94 -> 4.4 | Improved |
| use noise matrix alone without pseudo-label (ablation) | Radar-APLANC · Equipleth | MAE: 8.94 -> 34.48 | Hurt |
| add augmented pseudo-label stage two (ablation) | Radar-APLANC · Equipleth | MAE: 4.40 -> 3.95 | Improved |
| remove noise matrix in stage two while using augmented pseudo-label (ablation) | Radar-APLANC · Equipleth | MAE: 3.95 -> 7.42 | Hurt |
| add traditional signals and noise-based quality assessment to pretrained pseudo-label (ablation) | Radar-APLANC · Equipleth | MAE: 4.56 -> 3.95 | Improved |
| use traditional heartbeat signals with noise signal but no pretrained heartbeat signal (ablation) | Radar-APLANC · Equipleth | MAE: 3.95 -> 14.48 | Hurt |
| unsupervised training instead of supervised on cross-dataset RHB->Equipleth | Radar-APLANC · Equipleth (cross-dataset) | MAE: 4.10 vs 4.53 (Equipleth RF) | Improved |
| unsupervised vs supervised in-dataset Equipleth | Radar-APLANC · Equipleth | MAE: 3.95 vs 2.18 (Equipleth RF) | Hurt |
Closest papers by meaning
- PhysioWave: A Multi-Scale Wavelet-Transformer for Physiological Signal Representation (NeurIPS 2025 · similarity 0.80)
- Parallel-Learning of Invariant and Tempo-variant Attributes of Single-Lead Cardiac Signals: PLITA (AAAI 2025 · similarity 0.80)
- Cylindrical Geodesic Flow Matching for Quasiperiodic Physiological Signal Transformation (NeurIPS 2026 · similarity 0.79)
- AdaPKC: PeakConv with Adaptive Peak Receptive Field for Radar Semantic Segmentation (NeurIPS 2024 · similarity 0.79)
- NEAR: Neural Electromagnetic Array Response (ICML 2025 · similarity 0.78)
- FacePhys: State of the Heart Learning (NeurIPS 2026 · similarity 0.78)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
