Tag: multi-task learning

Researchers have developed HeteroGrad-DT to mitigate gradient interference and improve performance in heterogeneous multi-task offline reinforcement learning.

A retrospective study of 324 eyes demonstrates that a multi-task Bayesian Hierarchical Gated Transformer improves postoperative vault prediction after Implantable ...

A benchmark on 2,500 grocery households shows static features outperform deep recurrent sequences in predicting customer lifetime value, while marketing ...

Computer scientists have developed DTRec, a dual-triangular recommendation model that integrates large language models and knowledge graphs within a multi-task ...

A gradient-dependence-guided sparse learning framework reduces multi-task interference by combining Fisher information utility estimates with dynamic gradient trajectory tracking.

A new model called MIXTAP uses multi-task learning and knowledge distillation to jointly predict intra- and inter-protein PTM crosstalk, outperforming ...

A joint deep learning framework called J-SUP unifies aspect-based sentiment analysis and consumer preference profiling within a shared Transformer architecture ...

A single attention-gated multi-task network achieved classification and segmentation accuracy comparable to separate dedicated models on breast ultrasound images while ...

A dual-decoder deep-learning network localized 14 shoulder bony landmarks on contralateral CT scans with a mean radial error of 2.1 ...

UniLoopNet predicts enhancer–promoter, enhancer–enhancer, CTCF-mediated, and stripe-associated chromatin interactions from DNA sequence using a multi-task deep learning framework.