Tackling Multiple Tasks with One Single Learning Framework
Not on the map. Why?
Abstract (arXiv)
Deep Multi-Task Learning (DMTL) has been widely studied in the machine learning community and applied to a broad range of real-world applications. Searching for the optimal knowledge sharing in DMTL is more challenging for sequential learning problems, as the task relationship will change in the temporal dimension. In this paper, we propose a flexible and efficient framework called HierarchicalTemporal Activation Network (HTAN) to simultaneously explore the optimal sharing of the neural network hierarchy (hierarchical axis) and the time-variant task relationship (temporal axis). HTAN learns a set of time-variant activation functions to encode the task relation. A functional regularization implemented by a modulated SPDNet and adversarial learning is further proposed to enhance the DMTL performance. Comprehensive experiments on several challenging applications demonstrate that our HTAN-SPD framework outperforms SOTA methods significantly in sequential DMTL.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 30 of its 44 references that are in the map. First posted on arXiv 29 May 2022.
Builds on
30 of 44 references.
References · 30
- 2017Attention Is All You Needcited by 59,932
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,854
- 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,345
- 2018Multi-Task Learning as Multi-Objective Optimizationcited by 809
- 2018End-to-End Multi-Task Learning with Attentioncited by 533
- 2019Multi-Task Deep Neural Networks for Natural Language Understandingcited by 349
- 2016A Riemannian Network for SPD Matrix Learningcited by 162
- 2017Adversarial Multi-task Learning for Text Classificationcited by 91
- 2014Learning Activation Functions to Improve Deep Neural Networkscited by 62
- 2018A Hierarchical Multi-task Approach for Learning Embeddings from Semantic Taskscited by 54
- 2019An Interactive Multi-Task Learning Network for End-to-End Aspect-Based Sentiment Analysiscited by 52
18 more
- 2019Investigating Meta-Learning Algorithms for Low-Resource Natural Language Understanding Taskscited by 48
- 2019Answering while Summarizing: Multi-task Learning for Multi-hop QA with Evidence Extractioncited by 39
- 2020Fashion Captioning: Towards Generating Accurate Descriptions with Semantic Rewardscited by 37
- 2021ReFormer: The Relational Transformer for Image Captioningcited by 23
- 2019Multi-task Learning for Aggregated Data using Gaussian Processescited by 20
- 2018Meta Multi-Task Learning for Sequence Modelingcited by 18
- 2021Journalistic Guidelines Aware News Image Captioningcited by 14
- 2018Learning to Multitaskcited by 11
- 2018Multi-Task Learning with Multi-View Attention for Answer Selection and Knowledge Base Question Answeringcited by 11
- 2019Exploiting Entity BIO Tag Embeddings and Multi-task Learning for Relation Extraction with Imbalanced Datacited by 11
- 2019Multi-Task Networks With Universe, Group, and Task Feature Learningcited by 11
- 2019Different Absorption from the Same Sharing: Sifted Multi-task Learning for Fake News Detectioncited by 11
- 2020Crossing-Domain Generative Adversarial Networks for Unsupervised Multi-Domain Image-to-Image Translationcited by 9
- 2019Latent Part-of-Speech Sequences for Neural Machine Translationcited by 7
- 2020Learning Tuple Compatibility for Conditional OutfitRecommendationcited by 7
- 2019Recognizing License Plates in Real-Timecited by 4
- 2018GIRNet: Interleaved Multi-Task Recurrent State Sequence Modelscited by 3
- 2018Joint Multitask Learning for Community Question Answering Using Task-Specific Embeddingscited by 2
Built on by
Not cited yet.