Tag: autoencoders

Researchers in Stuttgart have introduced a machine-learning framework that identifies stable, passive port-Hamiltonian models directly from data, validated on systems ...

Researchers report a neighbor contrastive learning method that fills missing data views by transferring neighbor relationships and improves clustering accuracy ...

A study combining Transformer, deep neural-network and recurrent models with adaptive weighting reported 95% accuracy and 96% recall when detecting ...

Physicists have borrowed one of the most elegant constructions from quantum many-body physics and turned it into an anomaly detector ...

A transformer-autoencoder method that groups federated clients without labels improved sentiment-classification accuracy across controlled label-skew conditions on Sentiment140 and IMDB.

A new benchmark study shows that simple linear imputation methods outperform deep learning models when realistic spatio-temporal missingness is simulated ...

A systematic review maps how deep learning architectures are transforming EEG analysis for neurological disease diagnosis, while warning that data ...

A reproducibility study of 18 influential papers finds that data leakage, silently modified datasets, and underspecified baselines have inflated claims ...

A new open-source benchmark called towerbench brings frozen evaluation protocols, paired bootstrap statistics and rolling-origin windows to recommender system testing, ...

A hybrid systematic review and experimental benchmark of five deep learning architectures reveals which neural networks can realistically detect and ...