Contextualizing Multiple Tasks via Learning to Decompose
Lists
Only you see this
Abstract
One single instance could possess multiple portraits and reveal diverse relationships with others according to different contexts. Those ambiguities increase the difficulty of learning a generalizable model when there exists one concept or mixed concepts in a task. We propose a general approach Learning to Decompose Network (LeadNet) for both two cases, which contextualizes a model through meta-learning multiple maps for concepts discovery -- the representations of instances are decomposed and adapted conditioned on the contexts. Through taking a holistic view over multiple latent components over instances in a sampled pseudo task, LeadNet learns to automatically select the right concept via incorporating those rich semantics inside and between objects. LeadNet demonstrates its superiority in various applications, including exploring multiple views of confusing tasks, out-of-distribution recognition, and few-shot image classification.
BibTeX
@article{ye2024contextualizing,
author = {Han-Jia Ye and Da-Wei Zhou and Lanqing Hong and Zhenguo Li and Xiu-Shen Wei and De-Chuan Zhan},
title = {{Contextualizing Multiple Tasks via Learning to Decompose}},
journal = {IEEE Trans. Pattern Anal. Mach. Intell.},
year = {2024},
doi = {10.1109/tpami.2023.3317425},
}| Meta-Learning Amidst Heterogeneity and Ambiguity | arXiv · 2021 |
| Meta-learning from Tasks with Heterogeneous Attribute Spaces | NeurIPS · 2020 |
| Toward Multimodal Model-Agnostic Meta-Learning | arXiv · 2018 |
| HetMAML: Task-Heterogeneous Model-Agnostic Meta-Learning for Few-Shot Learning Across Modalities | arXiv · 2021 |
| Multimodal meta-learning through meta-learned task representations | Neural Computing and Applications · 2024 |
| Learning to Learn to Disambiguate: Meta-Learning for Few-Shot Word Sense Disambiguation | EMNLP · 2020 |