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

We develop novel methodology for active feature acquisition (AFA), the studyof how to sequentially acquire a dynamic (on a per instance basis) subset offeatures that minimizes acquisition costs whilst still yielding accuratepredictions. The AFA framework can be useful in a myriad of domains, includinghealth care applications where the cost of acquiring additional features for apatient (in terms of time, money, risk, etc.) can be weighed against theexpected improvement to diagnostic performance. Previous approaches for AFAhave employed either: deep learning RL techniques, which have difficultytraining policies in the AFA MDP due to sparse rewards and a complicated actionspace; deep learning surrogate generative models, which require modelingcomplicated multidimensional conditional distributions; or greedy policies,which fail to account for how joint feature acquisitions can be informativetogether for better predictions. In this work we show that we can bypass manyof these challenges with a novel, nonparametric oracle based approach, which wecoin the acquisition conditioned oracle (ACO). Extensive experiments show thesuperiority of the ACO to state-of-the-art AFA methods when acquiring featuresfor both predictions and general decision-making.

Acquisition Conditioned Oracle for Nongreedy Active Feature Acquisition