STDA-Net achieves 89.03% average accuracy and 87.64% macro F1 in cross-dataset sleep staging by processing 2D spectrograms with temporal modeling and unsupervised adversarial alignment, outperforming 1D baselines with lower variance.
Domain-adversarial training of neural networks
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 4verdicts
UNVERDICTED 4roles
background 1polarities
background 1representative citing papers
MOSAIC decomposes user intent into three orthogonal components via a triple-encoder architecture with adversarial training and dynamic gating to outperform baselines in multi-domain session recommendations.
Gen-ROTDA fits a target anchor with small labeled data, transfers residuals via robust OT with a generative feature generator and trims high-cost matches, achieving lowest MAE on 2025-2026 Citi Bike prediction and better stability than non-robust OT methods.
GPP trains local variational encoders in federated settings to release representations that keep utility within 1% of an autoencoder baseline while driving adversary AUC on sensitive attributes to near-random levels on MNIST, CelebA, and HAPT data.
citing papers explorer
-
STDA-Net: Spectrogram-Based Domain Adaptation for cross-dataset Sleep Stage Classification
STDA-Net achieves 89.03% average accuracy and 87.64% macro F1 in cross-dataset sleep staging by processing 2D spectrograms with temporal modeling and unsupervised adversarial alignment, outperforming 1D baselines with lower variance.
-
MOSAIC: Multi-Domain Orthogonal Session Adaptive Intent Capture for Prescient Recommendations
MOSAIC decomposes user intent into three orthogonal components via a triple-encoder architecture with adversarial training and dynamic gating to outperform baselines in multi-domain session recommendations.
-
Robust OT-Guided Generative Residual Domain Adaptation for Bike-Sharing Demand Prediction under Temporal Domain Shift
Gen-ROTDA fits a target anchor with small labeled data, transfers residuals via robust OT with a generative feature generator and trims high-cost matches, achieving lowest MAE on 2025-2026 Citi Bike prediction and better stability than non-robust OT methods.
-
Distributed Deep Variational Approach for Privacy-preserving Data Release
GPP trains local variational encoders in federated settings to release representations that keep utility within 1% of an autoencoder baseline while driving adversary AUC on sensitive attributes to near-random levels on MNIST, CelebA, and HAPT data.