Research map / Architectures and efficiency / RNNs and normalization
Fully-Adaptive Feature Sharing in Multi-Task Networks With Applications in Person Attribute Classification
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Automatically group similar tasks into shared branches instead of using a deliberately dissimilar grouping (ablation) | Branch-32-2.0 · CelebA | Per-task accuracy | Helped |
| Separate tasks that were grouped together in the original network (ablation) | Branch-32-2.0 · CelebA | Per-task accuracy | Mixed |
| Use pretrained initialization rather than training Branch-32-2.0 from scratch (ablation) | Branch-32-2.0 · CelebA | Accuracy gap vs VGG-16 Baseline: -0.54 vs -0.65 | No clear effect |
| Use pretrained initialization rather than training Branch-32-2.0 from scratch (ablation) | Branch-32-2.0 · CelebA | Top-10 Recall gap vs VGG-16 Baseline: -2.47 vs -3.77 | Helped |
| Initialize the thin network with SOMP-selected pretrained filters instead of not using SOMP initialization (ablation) | Baseline-thin-32 · CelebA | Training and validation accuracy over training | Helped |
| Add Batch Normalization to the thin baseline (ablation) | Baseline-thin-32 · CelebA | Training error | Helped |
| Jointly train facial and clothing attribute prediction in one compact network | Joint Branch-64-2.0 · CelebA and DeepFashion | Accuracy: 91.02 CelebA accuracy; 83.24 DeepFashion Top-3 Accuracy; 90.39 DeepFashion Top-5 Accuracy | Mixed |
Closest papers by meaning
- Attributes for Improved Attributes: A Multi-Task Network Utilizing Implicit and Explicit Relationships for Facial Attribute Classification (AAAI 2017 · similarity 0.88)
- Partially Shared Multi-Task Convolutional Neural Network With Local Constraint for Face Attribute Learning (CVPR 2018 · similarity 0.87)
- Deep Multi-task Representation Learning: A Tensor Factorisation Approach (ICLR 2017 · similarity 0.86)
- Distribution Matching for Multi-Task Learning of Classification Tasks: A Large-Scale Study on Faces & Beyond (AAAI 2024 · similarity 0.85)
- Multi-Task Dense Prediction via Mixture of Low-Rank Experts (CVPR 2024 · similarity 0.84)
- Provable Multi-Task Representation Learning by Two-Layer ReLU Neural Networks (ICML 2024 · similarity 0.83)
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.