AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Dynamic Task Prioritization for Multitask Learning

Michelle Guo, Albert Haque, De-An Huang, Serena Yeung, Li Fei-Fei

ECCV 2018 · Architectures and efficiency › RNNs and normalization · approach: Dataset distillation and data-efficient training · 395 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
replace uniform task weights with dynamic task prioritization (hard-first)hard parameter sharing DenseNet · COCO+MPII multitask (hard sharing, task weighting)pose AP50: 47.72 -> 56.11Improved
prioritize hard tasks instead of easy tasks in dynamic priority (ablation)dynamic task prioritization · COCO+MPII multitask (hard sharing, task weighting)pose AP50: 32.90 -> 56.11Improved
use dynamic hard-task priority instead of self-paced weightingmultitask DenseNet · COCO+MPII multitask (hard sharing, task weighting)pose AP50: 49.34 -> 56.11Improved
add a learning rate schedule to fixed Prioritize Hard weights (ablation)Prioritize Hard · COCO+MPII multitask (hard sharing, task weighting)detection AP50: 65.86 -> 82.64Improved
add example-level focal loss (gamma=1) to uniform weighting (ablation)uniform multitask loss · COCO+MPII multitask (hard sharing, task weighting)segmentation AP50: pose 47.72 -> 54.04; segmentation 78.57 -> 70.77Mixed
increase task-level focusing parameter from 1 to 2 (ablation)DTP · COCO+MPII multitask (hard sharing, task weighting)classification top-1: 91.1 -> 88.8Hurt
increase example-level focusing parameter gamma0 from 1 to 2 (ablation)FL · COCO+MPII multitask (hard sharing, task weighting)detection AP50: 82.0 -> 69.9Hurt
add example-level focal loss (gamma0=1) on top of DTP (ablation)DTP · COCO+MPII multitask (hard sharing, task weighting)detection AP50: detection 79.3 -> 85.9; classification 91.1 -> 88.9Mixed
replace hard parameter sharing with a task hierarchy ordered by difficulty (ablation)DenseNet multitask · COCO multitask architecture comparisonpose AP50: 56.11 -> 62.40Improved
replace UberNet with task hierarchy (ablation)DenseNet multitask · COCO multitask architecture comparisonsegmentation AP50: 77.81 -> 80.26Improved
place segmentation in the middle of the task hierarchy rather than first (ablation)task hierarchy · COCO segmentationrelative performance: boosted by 1.25Improved
train all four tasks jointly instead of single-task detection modeldynamic task prioritization · COCO detectionAP: 49.4 -> 52.5Improved

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