SensorFault-Bench is a new CPS-grounded benchmark showing that clean-MSE rankings of forecasting models often disagree with their robustness under standardized sensor-fault scenarios across four real datasets.
An empirical survey of data augmentation for time series classification with neural networks.PLOS ONE, 16(7):e0254841
6 Pith papers cite this work, alongside 671 external citations. Polarity classification is still indexing.
6
Pith papers citing it
671
external citations · Crossref
citation-role summary
background 2
method 1
citation-polarity summary
representative citing papers
A compact 175,685-parameter ResNet classifies 21 simulated 400 Hz aerospace fault and power-quality classes with 96.94% software accuracy and 95.87% accuracy after 8-bit quantization on a ZCU102 FPGA, at 6.90 ms mean accelerator latency.
A contrastive-learning framework with full-covariance 'angular Gaussian' class scoring plus adaptive logit adjustment reports the best macro accuracy for long-tailed multi-label ECG arrhythmia diagnosis on PTB-XL and a new nocturnal cohort.
Sonata is a small hybrid world model pre-trained to predict future IMU states that outperforms autoregressive baselines on clinical discrimination, fall-risk prediction, and cross-cohort transfer while fitting on-device wearables.
A knowledge-distilled accelerometry encoder, taught by an unsupervised PPG teacher on 20 million minutes of paired wearable data, predicts heart rate, heart-rate variability, demographics, and 46 health conditions from motion alone.
A transformer-based variational autoencoder applies time series augmentations in its latent space, claiming better control and fidelity than direct augmentation, but the reported Wasserstein results contradict that claim on one of three datasets.
citing papers explorer
-
Benchmarking Sensor-Fault Robustness in Forecasting
SensorFault-Bench is a new CPS-grounded benchmark showing that clean-MSE rankings of forecasting models often disagree with their robustness under standardized sensor-fault scenarios across four real datasets.
-
Angular Gaussian Supervised Contrastive Learning for Long-Tailed Electrocardiogram Arrhythmia Diagnosis
A contrastive-learning framework with full-covariance 'angular Gaussian' class scoring plus adaptive logit adjustment reports the best macro accuracy for long-tailed multi-label ECG arrhythmia diagnosis on PTB-XL and a new nocturnal cohort.
-
Sonata: A Hybrid World Model for Inertial Kinematics under Clinical Data Scarcity
Sonata is a small hybrid world model pre-trained to predict future IMU states that outperforms autoregressive baselines on clinical discrimination, fall-risk prediction, and cross-cohort transfer while fitting on-device wearables.
-
Wearable Accelerometer Foundation Models for Health via Knowledge Distillation
A knowledge-distilled accelerometry encoder, taught by an unsupervised PPG teacher on 20 million minutes of paired wearable data, predicts heart rate, heart-rate variability, demographics, and 46 health conditions from motion alone.
-
L-GTA: Latent Generative Modeling for Time Series Augmentation
A transformer-based variational autoencoder applies time series augmentations in its latent space, claiming better control and fidelity than direct augmentation, but the reported Wasserstein results contradict that claim on one of three datasets.