AI Paper Atlas

Research map / Image generation / Super-resolution

Spatial Projection of Multiple Climate Variables Using Hierarchical Multitask Learning

Andre Goncalves, Arindam Banerjee, Fernando Von Zuben

AAAI 2017 · Image generation › Super-resolution · approach: Perceptual and real-world SR · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Couple climate-variable MTL problems using shared sparsity patterns in their precision matricesHMTL · South American summer precipitation, 20 training yearsRMSE: 7.16 (0.43) -> 6.48 (0.34)Improved
Couple climate-variable MTL problems using shared sparsity patterns in their precision matricesHMTL · South American summer precipitation, 30 training yearsRMSE: 6.86 (0.48) -> 6.37 (0.29)Improved
Couple climate-variable MTL problems using shared sparsity patterns in their precision matricesHMTL · South American yearly temperature, 20 training yearsRMSE: 1.00 (0.04) -> 0.91 (0.02)Improved
Increase the number of super-tasks from 2 to 7HMTL · Synthetic data, 50 training samplesrelative RMSE improvement over MSSL: 2.2 -> 5.8Improved
Increase the number of super-tasks from 2 to 7HMTL · Synthetic data, 30 training samplesrelative RMSE improvement over MSSL: 22.3 -> 24.3Improved
Use equal weights for all ESMs instead of a learned ensembleClimate projection · South American precipitation and temperature scenariosRMSEHurt
Use graph-Laplacian spatial smoothing with predefined neighborhood dependenceS²M²R · Yearly climate dataRMSEHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy