Rethinking Hard-Parameter Sharing in Multi-Domain Learning

Zhang, Lijun; Yang, Qizheng; Liu, Xiao; Guan, Hui

doi:10.1109/icme52920.2022.9859706

Cited by 9 publications

(8 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NumDomain + = 1 layers. We know from prior research that the end layer of a CNN has lower representation capacity compared to other layers in the architecture and is thus more sensitive to domain-specific information 43 . We, therefore, choose n to be the last fully connected layer of the CNN architecture shown in Fig.…”

Section: Federated Multi-domain Learningmentioning

confidence: 99%

Federated Clustered Multi-Domain Learning for Health Monitoring

Jiang,

Li,

Firouzi

et al. 2024

Preprint

View full text Add to dashboard Cite

Wearable Internet of Things (WIoT) and Artificial Intelligence (AI) are rapidly emerging technologies for healthcare. These technologies enable seamless data collection and precise analysis toward fast, resource-abundant, and personalized patient care. However, conventional machine learning workflow requires data to be transferred to the remote cloud server, which leads to significant privacy concerns. To tackle this problem, researchers have proposed federated learning, where end-point users collaboratively learn a shared model without sharing local data. However, data heterogeneity, i.e., variations in data distributions within a client (intra-client) or across clients (inter-client), degrades the performance of federated learning. Existing state-of-the-art methods mainly consider inter-client data heterogeneity, whereas intra-client variations have not received much attention. To address intra-client variations in federated learning, we propose a federated clustered multi-domain learning algorithm based on ClusterGAN, multi-domain learning, and graph neural networks. We applied the proposed algorithm to a case study on stress-level prediction, and our proposed algorithm outperforms two state-of-the-art methods by 4.4% in accuracy and 0.06 in the F1 score. In addition, we demonstrate the effectiveness of the proposed algorithm by investigating variants of its different modules.

show abstract

Section: Federated Multi-domain Learningmentioning

confidence: 99%

Federated Clustered Multi-Domain Learning for Health Monitoring

Jiang,

Li,

Firouzi

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Multi-task learning [21][22][23] is the joint training of multiple related tasks, aiming to improve the generalization and robustness of each task through information transmission and integration among multiple tasks. Naturally multi-task learning is more in line with human cognitive learning mechanism.…”

Section: Multi-task Learningmentioning

confidence: 99%

ContourNet:Research on Contour Based Nighttime Semantic Segmentation

Yang

Liu

2023

Preprint

View full text Add to dashboard Cite

Due to the scarcity of nighttime semantic segmentation datasets and the high demand for network models, the development of semantic segmentation of nighttime scenes is still very slow. This paper proposes a new network model, ContourNet, which can model images in multiple stages, including local features with rich superficial appearance information, global features with rich deep semantic information, and intermediate layer features. At the same time, this paper uses multi head decoders to fuse different levels of features. In this paper, a separate contour network module for processing object contours is designed for night images with imperceptible texture and color information. This network module is parallel to the semantic network module to achieve accurate prediction of object contours. The performance is prominent, especially for objects that are far away, small objects, or objects with high contour continuity. A large number of experiments demonstrates that the ContourNet proposed in this paper can significantly improve the semantic segmentation ability of existing models for nighttime images, and can also improve the semantic segmentation accuracy of daytime images to a certain extent, with good generalization ability. Specifically, after adding the contour module in this article, MIoU has increased by 5.1% on the night dataset Rebecca; MIoU has increased by 2.5% on the daytime dataset CamVid.

show abstract

“…Hard parameter sharing is the most commonly used approach by which the hidden CNN layers are shared between all tasks while keeping several task-specific outputs (Zhang et al, 2022). Caruana (1997) present examples of the usages of MTL.…”

Section: Background and Related Workmentioning

confidence: 99%

Hierarchical Classification of Insects with Multitask Learning and Anomaly Detection

Bjerge¹,

Geissmann

Alison

et al. 2023

Preprint

View full text Add to dashboard Cite

Cameras and computer vision are revolutionising the study of insects, creating new research opportunities within agriculture, epidemiology, evolution, ecology and monitoring of biodiversity. However, a major challenge is the diversity of insects and close resemblances of many species combined with computer vision are often not sufficient to classify large numbers of insect species, which sometimes cannot be identified at the species level. Here, we present an algorithm to hierarchically classify insects from images, leveraging a simple taxonomy to (1) classify specimens across multiple taxonomic ranks simultaneously, and (2) highlight the lowest rank at which a reliable classification can be reached. Specifically, we propose multitask learning, a loss function incorporating class dependency at each taxonomic rank, and anomaly detection based on outlier analysis for quantification of uncertainty. First, we compile a dataset of 41,731 images of insects, combining images from time-lapse monitoring of floral scenes with images from the Global Biodiversity Information Facility (GBIF). Second, we adapt state-of-the-art convolutional neural networks, ResNet and EfficientNet, for the hierarchical classification of insects belonging to three orders, five families and nine species. Third, we assess model generalization for 11 species unseen by the trained models. Here, anomaly detection is used to predict the higher rank of the species not present in the training set. We found that incorporating a simple taxonomy into our model increased accuracy at higher taxonomic ranks. As expected, our algorithm correctly classified new insect species at higher taxonomic ranks, while classification was uncertain at lower taxonomic ranks. Anomaly detection can effectively flag novel taxa that are visually distinct from species in the training data. However, five novel taxa were consistently mistaken for visually similar species in the training data. Above all, we have demonstrated a practical approach to hierarchical classification based on species taxonomy and uncertainty during automated in situ monitoring of live insects. Our method is simple and versatile and could be implemented to classify a wide range of insects as well as other organisms.

show abstract

Rethinking Hard-Parameter Sharing in Multi-Domain Learning

Cited by 9 publications

References 5 publications

Federated Clustered Multi-Domain Learning for Health Monitoring

Federated Clustered Multi-Domain Learning for Health Monitoring

ContourNet:Research on Contour Based Nighttime Semantic Segmentation

Hierarchical Classification of Insects with Multitask Learning and Anomaly Detection

Contact Info

Product

Resources

About