What and Where: Learn to Plug Adapters via NAS for Multi-Domain Learning

2020/07/24 by Hanbin Zhao, Hao Zeng, Zhao, Hanbin +11
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and ELM #Multimodal Machine Learning Applications

paper · pdf · doi:10.48550/arxiv.2007.12415

openalex publication_date 2020/07/24 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

As an important and challenging problem, multi-domain learning (MDL) typically seeks for a set of effective lightweight domain-specific adapter modules plugged into a common domain-agnostic network. Usually, existing ways of adapter plugging and structure design are handcrafted and fixed for all domains before model learning, resulting in the learning inflexibility and computational intensiveness. With this motivation, we propose to learn a data-driven adapter plugging strategy with Neural Architecture Search (NAS), which automatically determines where to plug for those adapter modules. Furthermore, we propose a NAS-adapter module for adapter structure design in a NAS-driven learning scheme, which automatically discovers effective adapter module structures for different domains. Experimental results demonstrate the effectiveness of our MDL model against existing approaches under the conditions of comparable performance.

Citations

• ImageNet Large Scale Visual Recognition Challenge
• Learning without Forgetting
• Designing Neural Network Architectures using Reinforcement Learning
• Universal representations:The missing link between faces, text, planktons, and cat breeds
• DARTS: Differentiable Architecture Search
• ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
• Feature Partitioning for Efficient Multi-Task Architectures
• PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search
• RC-DARTS: Resource Constrained Differentiable Architecture Search
• Learning Multiple Layers of Features from Tiny Images

Related

• Cascaded Multi-task Adaptive Learning Based on Neural Architecture Search
• Network Architecture Search for Domain Adaptation
• ModuleNet: Knowledge-inherited Neural Architecture Search
• Learnable Embedding Space for Efficient Neural Architecture Compression
• CLOSE: Curriculum Learning On the Sharing Extent Towards Better One-shot NAS