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

Multi-task learning (MTL) aims to improve the performance of a primary taskby jointly learning with related auxiliary tasks. Traditional MTL methodsselect tasks randomly during training. However, both previous studies and ourresults suggest that such a random selection of tasks may not be helpful, andcan even be harmful to performance. Therefore, new strategies for taskselection and assignment in MTL need to be explored. This paper studies themulti-modal, multi-task dialogue act classification task, and proposes a methodfor selecting and assigning tasks based on non-stationary multi-armed bandits(MAB) with discounted Thompson Sampling (TS) using Gaussian priors. Ourexperimental results show that in different training stages, different taskshave different utility. Our proposed method can effectively identify the taskutility, actively avoid useless or harmful tasks, and realise the taskassignment during training. Our proposed method is significantly superior interms of UAR and F1 to the single-task and multi-task baselines with p-values <0.05. Further analysis of experiments indicates that for the dataset with thedata imbalance problem, our proposed method has significantly higher stabilityand can obtain consistent and decent performance for minority classes. Ourproposed method is superior to the current state-of-the-art model.

Task Selection and Assignment for Multi-modal Multi-task Dialogue Act Classification with Non-stationary Multi-armed Bandits