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

Abstract

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.

A Distributed Aggregation Approach for Vehicular Federated Learning

“…The KF is an integral part of the contact estimation process. This article assumes KF as a mobility prediction mechanism, but other approaches could be used, such as the work by Emami et al [36].…”

INTRAFORCE: Intra-Cluster Reinforced Social Transformer for Trajectory Prediction

“…In recent years, data-driven social mobility predictors are gaining popularity compared to the previously proposed Social-force models, which use simple repulsive and attraction forces [7]. The vast majority of modern human-trajectory predictors are based on deep learning models, such as RNNs, Long Short-Term Memorys (LSTMs), Convolutional Neural Networks (CNNs), and attention-based neural networks, such as Transformers, which require less computation and achieve higher prediction accuracy compared to social-force models due to their better modeling of sequential patterns [1], [8], [9]. Instead of modeling kinetic forces and energy potentials as in social-force models, social-pooling [2], [3], attention [10], [5],…”

“…Hence, the absence of RL search mechanisms in less-explored fields is apparent [12]. RC-TL [8] suggests applying RL for individual trajectory prediction, where users are left isolated. In this direction, INTRAFORCE proposes and studies a RL-based NAS method in the field of social trajectory prediction.…”