CrowdTransfer: Enabling Crowd Knowledge Transfer in AIoT Community
arXiv:2407.06485 · doi:10.1109/COMST.2024.3423319
Abstract
Artificial Intelligence of Things (AIoT) is an emerging frontier based on the deep fusion of Internet of Things (IoT) and Artificial Intelligence (AI) technologies. Although advanced deep learning techniques enhance the efficient data processing and intelligent analysis of complex IoT data, they still suffer from notable challenges when deployed to practical AIoT applications, such as constrained resources, and diverse task requirements. Knowledge transfer is an effective method to enhance learning performance by avoiding the exorbitant costs associated with data recollection and model retraining. Notably, although there are already some valuable and impressive surveys on transfer learning, these surveys introduce approaches in a relatively isolated way and lack the recent advances of various knowledge transfer techniques for AIoT field. This survey endeavors to introduce a new concept of knowledge transfer, referred to as Crowd Knowledge Transfer (CrowdTransfer), which aims to transfer prior knowledge learned from a crowd of agents to reduce the training cost and as well as improve the performance of the model in real-world complicated scenarios. Particularly, we present four transfer modes from the perspective of crowd intelligence, including derivation, sharing, evolution and fusion modes. Building upon conventional transfer learning methods, we further delve into advanced crowd knowledge transfer models from three perspectives for various AIoT applications. Furthermore, we explore some applications of AIoT areas, such as human activity recognition, urban computing, multi-robot system, and smart factory. Finally, we discuss the open issues and outline future research directions of knowledge transfer in AIoT community.
This paper has been accepted for publication in IEEE Communications Surveys & Tutorials. Copyright will be transferred without notice, after this version may no longer be accessible
References in corpus (71)
- Deep Learning in Neural Networks: An Overview
- Distilling the Knowledge in a Neural Network
- Overcoming catastrophic forgetting in neural networks
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Training language models to follow instructions with human feedback
- Learning Transferable Features with Deep Adaptation Networks
- Unsupervised Domain Adaptation by Backpropagation
- LoRA: Low-Rank Adaptation of Large Language Models
- FitNets: Hints for Thin Deep Nets
- Dynamic reconfiguration of human brain networks during learning
- Federated Learning for Internet of Things: A Comprehensive Survey
- Deep Multi-modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges
- Edge Intelligence: The Confluence of Edge Computing and Artificial Intelligence
- Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
- Secure Federated Transfer Learning
- Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization
- Efficient Lifelong Learning with A-GEM
- Deep Learning for Image and Point Cloud Fusion in Autonomous Driving: A Review
- 3D-CVF: Generating Joint Camera and LiDAR Features Using Cross-View Spatial Feature Fusion for 3D Object Detection
- Ensemble Distillation for Robust Model Fusion in Federated Learning
- FedMD: Heterogenous Federated Learning via Model Distillation
- Born Again Neural Networks
- Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning
- Adversarial Learning for Semi-Supervised Semantic Segmentation
- Large-Scale Study of Curiosity-Driven Learning
- Lifelong Federated Reinforcement Learning: A Learning Architecture for Navigation in Cloud Robotic Systems
- Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning
- Federated Learning in the Sky: Aerial-Ground Air Quality Sensing Framework with UAV Swarms
- One-Shot Imitation Learning
- Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentation
- Interactive Policy Learning through Confidence-Based Autonomy
- Encoder Based Lifelong Learning
- Federated Learning with Bayesian Differential Privacy
- Group Knowledge Transfer: Federated Learning of Large CNNs at the Edge
- Learning Multiple Tasks with Multilinear Relationship Networks
- Decentralized Federated Learning through Proxy Model Sharing
- Unsupervised Transfer Learning for Anomaly Detection: Application to Complementary Operating Condition Transfer
- Meta-HAR: Federated Representation Learning for Human Activity Recognition
- Federated Transfer Learning for EEG Signal Classification
- LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters
- Transferability in Deep Learning: A Survey
- Feature-map-level Online Adversarial Knowledge Distillation
- LEEP: A New Measure to Evaluate Transferability of Learned Representations
- Efficient Test-Time Model Adaptation without Forgetting
- AIM: Adapting Image Models for Efficient Video Action Recognition
- Self-Supervised Aggregation of Diverse Experts for Test-Agnostic Long-Tailed Recognition
- Compacting Deep Neural Networks for Internet of Things: Methods and Applications
- Adversarial adaptive 1-D convolutional neural networks for bearing fault diagnosis under varying working condition
- DQN-TAMER: Human-in-the-Loop Reinforcement Learning with Intractable Feedback
- Enhancing Diversity in Teacher-Student Networks via Asymmetric branches for Unsupervised Person Re-identification
- Routing Networks and the Challenges of Modular and Compositional Computation
- Efficient Teacher: Semi-Supervised Object Detection for YOLOv5
- Multimodal Transfer Deep Learning with Applications in Audio-Visual Recognition
- Quantifying the Performance of Federated Transfer Learning
- GFL: A Decentralized Federated Learning Framework Based On Blockchain
- A Survey on Decentralized Federated Learning
- End-to-end Autonomous Driving: Challenges and Frontiers
- Generalized Proximal Policy Optimization with Sample Reuse
- Doing More with Less: Overcoming Data Scarcity for POI Recommendation via Cross-Region Transfer
- SparCL: Sparse Continual Learning on the Edge
- Catastrophic Interference in Reinforcement Learning: A Solution Based on Context Division and Knowledge Distillation
- SwiftQueue: Optimizing Low-Latency Applications with Swift Packet Queuing
- NISPA: Neuro-Inspired Stability-Plasticity Adaptation for Continual Learning in Sparse Networks
- TLeague: A Framework for Competitive Self-Play based Distributed Multi-Agent Reinforcement Learning
- Decentralized Federated Learning for UAV Networks: Architecture, Challenges, and Opportunities
- Multi-Agent Policy Transfer via Task Relationship Modeling
- Continual Learning and Private Unlearning
- Could robots be regarded as humans in future?
- Transferability-based Chain Motion Mapping from Humans to Humanoids for Teleoperation
- Beyond Invariance: Test-Time Label-Shift Adaptation for Distributions with "Spurious" Correlations