A Survey on Negative Transfer

Abstract

Transfer learning (TL) utilizes data or knowledge from one or more source domains to facilitate the learning in a target domain. It is particularly useful when the target domain has very few or no labeled data, due to annotation expense, privacy concerns, etc. Unfortunately, the effectiveness of TL is not always guaranteed. Negative transfer (NT), i.e., leveraging source domain data/knowledge undesirably reduces the learning performance in the target domain, has been a long-standing and challenging problem in TL. Various approaches have been proposed in the literature to handle it. However, there does not exist a systematic survey on the formulation of NT, the factors leading to NT, and the algorithms that mitigate NT. This paper fills this gap, by first introducing the definition of NT and its factors, then reviewing about fifty representative approaches for overcoming NT, according to four categories: secure transfer, domain similarity estimation, distant transfer, and NT mitigation. NT in related fields, e.g., multi-task learning, lifelong learning, and adversarial attacks, are also discussed.

References in corpus (12)

• Deep Subdomain Adaptation Network for Image Classification
• Applications of Unsupervised Deep Transfer Learning to Intelligent Fault Diagnosis: A Survey and Comparative Study
• Massively Multilingual Neural Machine Translation in the Wild: Findings and Challenges
• Switching EEG Headsets Made Easy: Reducing Offline Calibration Effort Using Active Weighted Adaptation Regularization
• Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Models
• Domain Adaptation by Class Centroid Matching and Local Manifold Self-Learning
• Tiny noise, big mistakes: Adversarial perturbations induce errors in Brain-Computer Interface spellers
• Learning Transferable Parameters for Unsupervised Domain Adaptation
• Source-free Domain Adaptation via Distributional Alignment by Matching Batch Normalization Statistics
• Newer is not always better: Rethinking transferability metrics, their peculiarities, stability and performance
• Headless Horseman: Adversarial Attacks on Transfer Learning Models
• Continuous Transfer Learning with Label-informed Distribution Alignment

Cited by in corpus (16)

• Transfer Learning for Motor Imagery Based Brain-Computer Interfaces: A Complete Pipeline
• All in One and One for All: A Simple yet Effective Method towards Cross-domain Graph Pretraining
• HetGPT: Harnessing the Power of Prompt Tuning in Pre-Trained Heterogeneous Graph Neural Networks
• DACAD: Domain Adaptation Contrastive Learning for Anomaly Detection in Multivariate Time Series
• Sharing to learn and learning to share; Fitting together Meta-Learning, Multi-Task Learning, and Transfer Learning: A meta review
• Enhanced Cross-Dataset Electroencephalogram-based Emotion Recognition using Unsupervised Domain Adaptation
• Amplifying Pathological Detection in EEG Signaling Pathways through Cross-Dataset Transfer Learning
• Pay Less But Get More: A Dual-Attention-based Channel Estimation Network for Massive MIMO Systems with Low-Density Pilots
• Revisiting Euclidean Alignment for Transfer Learning in EEG-Based Brain-Computer Interfaces
• Bridging the Gap: From Ad-hoc to Proactive Search in Conversations
• Cross-Task Inconsistency Based Active Learning (CTIAL) for Emotion Recognition
• Robust and Explainable Framework to Address Data Scarcity in Diagnostic Imaging
• Semi-Supervised Transfer Boosting (SS-TrBoosting)
• Energy-Efficient Prediction in Textile Manufacturing: Enhancing Accuracy and Data Efficiency With Ensemble Deep Transfer Learning
• Domain-Division based Progressive Learning for Source-Free Domain Adaptation
• Covariate-Elaborated Robust Partial Information Transfer with Conditional Spike-and-Slab Prior