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

Advances in object recognition flourished in part because of the availabilityof high-quality datasets and associated benchmarks. However, thesebenchmarks---such as ILSVRC---are relatively task-specific, focusingpredominately on predicting class labels. We introduce a publicly-availabledataset that embodies the task-general capabilities of human perception andreasoning. The Human Similarity Judgments extension to ImageNet (ImageNet-HSJ)is composed of human similarity judgments that supplement the ILSVRC validationset. The new dataset supports a range of task and performance metrics,including the evaluation of unsupervised learning algorithms. We demonstratetwo methods of assessment: using the similarity judgments directly and using apsychological embedding trained on the similarity judgments. This embeddingspace contains an order of magnitude more points (i.e., images) than previousefforts based on human judgments. Scaling to the full 50,000 image set was madepossible through a selective sampling process that used variational Bayesianinference and model ensembles to sample aspects of the embedding space thatwere most uncertain. This methodological innovation not only enables scaling,but should also improve the quality of solutions by focusing sampling where itis needed. To demonstrate the utility of ImageNet-HSJ, we used the similarityratings and the embedding space to evaluate how well several popular modelsconform to human similarity judgments. One finding is that more complex modelsthat perform better on task-specific benchmarks do not better conform to humansemantic judgments. In addition to the human similarity judgments, pre-trainedpsychological embeddings and code for inferring variational embeddings are madepublicly available. Collectively, ImageNet-HSJ assets support the appraisal ofinternal representations and the development of more human-like models.

Enriching ImageNet with Human Similarity Judgments and Psychological Embeddings