English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

As an effective learning paradigm against insufficient training samples,Multi-Task Learning (MTL) encourages knowledge sharing across multiple relatedtasks so as to improve the overall performance. In MTL, a major challengesprings from the phenomenon that sharing the knowledge with dissimilar and hardtasks, known as negative transfer, often results in a worsened performance.Though a substantial amount of studies have been carried out against thenegative transfer, most of the existing methods only model the transferrelationship as task correlations, with the transfer across features and tasksleft unconsidered. Different from the existing methods, our goal is toalleviate negative transfer collaboratively across features and tasks. To thisend, we propose a novel multi-task learning method called Task-FeatureCollaborative Learning (TFCL). Specifically, we first propose a base model witha heterogeneous block-diagonal structure regularizer to leverage thecollaborative grouping of features and tasks and suppressing inter-groupknowledge sharing. We then propose an optimization method for the model.Extensive theoretical analysis shows that our proposed method has the followingbenefits: (a) it enjoys the global convergence property and (b) it provides ablock-diagonal structure recovery guarantee. As a practical extension, weextend the base model by allowing overlapping features and differentiating thehard tasks. We further apply it to the personalized attribute predictionproblem with fine-grained modeling of user behaviors. Finally, experimentalresults on both simulated dataset and real-world datasets demonstrate theeffectiveness of our proposed method

Task-Feature Collaborative Learning with Application to Personalized Attribute Prediction