RC-TL: Reinforcement Convolutional Transfer Learning for Large-scale Trajectory Prediction

Emami, Negar; Pacheco, Lucas; Di Maio, Antonio; Braun, Torsten (9 June 2022). RC-TL: Reinforcement Convolutional Transfer Learning for Large-scale Trajectory Prediction. In: IEEE/IFIP Network Operations and Management Symposium (NOMS 2022) (pp. 1-9). IEEE Xplore: IEEE 10.1109/NOMS54207.2022.9789883

[img]
Preview
Text
Emami_RC_TL_preprint.pdf - Accepted Version
Available under License Publisher holds Copyright.

Download (1MB) | Preview
[img] Text
Emami_RC_TL.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (1MB) | Request a copy

Anticipating future locations of mobile users plays a pivotal role in intelligent services supporting mobile networks. Predicting user trajectories is a crucial task not only from the perspective of facilitating smart cities but also of significant importance in network management, such as handover optimization, service migration, and the caching of services in a mobile and edge-computing network. Convolutional Neural Networks (CNNs) have proven to be successful to tackle the forecasting of mobile users’ future locations. However, designing effective CNN architectures is challenging due to their large hyper-parameter space. Reinforcement Learning (RL)-based Neural Architecture Search (NAS) mechanisms have been proposed to optimize the neural network design process, but they are computationally expensive and they have not been used to predict user mobility. In large urban scenarios, the rate at which mobility information is generated makes it a challenge to optimize, train, and maintain prediction models for individual users. However, considering that user trajectories are not independent, a common trajectory-prediction model can be built and shared among a set of users characterized by similar mobility features. In the present work, we introduce Reinforcement Convolutional Transfer Learning (RC-TL), a CNN-based trajectory-prediction system that clusters users with similar trajectories, dedicates a single RL agent per cluster to optimize a CNN neural architecture, trains one model per cluster using the data of a small user subset, and transfers it to the other users in the cluster. Experimental results on a large-scale dataset show that our proposed RL-based CNN achieves up to 12% higher trajectory-prediction accuracy, with no training speed reduction, over other state-of-the-art approaches on a large-scale, real-world mobility dataset. Moreover, RC-TL’s clustering strategy saves up to 90% of the computational resources needed for training compared to single-user models, in exchange for a 3% accuracy reduction.

Item Type:

Conference or Workshop Item (Paper)

Division/Institute:

08 Faculty of Science > Institute of Computer Science (INF)
08 Faculty of Science > Institute of Computer Science (INF) > Communication and Distributed Systems (CDS)

UniBE Contributor:

Emami, Negar, de Sousa Pacheco, Lucas, Di Maio, Antonio, Braun, Torsten

Subjects:

000 Computer science, knowledge & systems
500 Science > 510 Mathematics

ISSN:

2374-9709

ISBN:

978-1-6654-0601-7

Publisher:

IEEE

Language:

English

Submitter:

Dimitrios Xenakis

Date Deposited:

23 Mar 2022 16:11

Last Modified:

27 Aug 2023 02:43

Publisher DOI:

10.1109/NOMS54207.2022.9789883

Uncontrolled Keywords:

Trajectory Prediction; Convolutional Neural Network, Reinforcement Learning; Clustering; Transfer Learning

BORIS DOI:

10.48350/166954

URI:

https://boris.unibe.ch/id/eprint/166954

Actions (login required)

Edit item Edit item
Provide Feedback