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

Many machine learning algorithms are trained and evaluated by splitting datafrom a single source into training and test sets. While such focus onin-distribution learning scenarios has led to interesting advancement, it hasnot been able to tell if models are relying on dataset biases as shortcuts forsuccessful prediction (e.g., using snow cues for recognising snowmobiles),resulting in biased models that fail to generalise when the bias shifts to adifferent class. The cross-bias generalisation problem has been addressed byde-biasing training data through augmentation or re-sampling, which are oftenprohibitive due to the data collection cost (e.g., collecting images of asnowmobile on a desert) and the difficulty of quantifying or expressing biasesin the first place. In this work, we propose a novel framework to train ade-biased representation by encouraging it to be different from a set ofrepresentations that are biased by design. This tactic is feasible in manyscenarios where it is much easier to define a set of biased representationsthan to define and quantify bias. We demonstrate the efficacy of our methodacross a variety of synthetic and real-world biases; our experiments show thatthe method discourages models from taking bias shortcuts, resulting in improvedgeneralisation. Source code is available at https://github.com/clovaai/rebias.

Learning De-biased Representations with Biased Representations