Long-term traffic prediction is highly challenging due to the complexity of traffic systems and the constantly changing nature of many impacting factors. In this paper, we focus on the spatio-temporal factors, and propose a graph multi-attention network (GMAN) to predict traffic conditions for time steps ahead at different locations on a road network graph. GMAN adapts an encoder-decoder architecture, where both the encoder and the decoder consist of multiple spatio-temporal attention blocks to model the impact of the spatio-temporal factors on traffic conditions. The encoder encodes the input traffic features and the decoder predicts the output sequence. Between the encoder and the decoder, a transform attention layer is applied to convert the encoded traffic features to generate the sequence representations of future time steps as the input of the decoder. The transform attention mechanism models the direct relationships between historical and future time steps that helps to alleviate the error propagation problem among prediction time steps. Experimental results on two real-world traffic prediction tasks (i.e., traffic volume prediction and traffic speed prediction) demonstrate the superiority of GMAN. In particular, in the 1 hour ahead prediction, GMAN outperforms state-of-the-art methods by up to 4% improvement in MAE measure. The source code is available at https://github.com/zhengchuanpan/GMAN.

Fairness has emerged as a critical problem in federated learning (FL). In this work, we identify a cause of unfairness in FL -- conflicting gradients with large differences in the magnitudes. To address this issue, we propose the federated fair averaging (FedFV) algorithm to mitigate potential conflicts among clients before averaging their gradients. We first use the cosine similarity to detect gradient conflicts, and then iteratively eliminate such conflicts by modifying both the direction and the magnitude of the gradients. We further show the theoretical foundation of FedFV to mitigate the issue conflicting gradients and converge to Pareto stationary solutions. Extensive experiments on a suite of federated datasets confirm that FedFV compares favorably against state-of-the-art methods in terms of fairness, accuracy and efficiency. The source code is available at https://github.com/WwZzz/easyFL.

No abstract

Abstract-The fast growing mobile network data traffic poses great challenges for operators to increase their data processing capacity in base stations in an efficient manner. With the emergence of Cloud Radio Access Network (Cloud-RAN), the data processing units can now be centralized in a data center and shared among several base stations. By clustering base stations with complementary traffic patterns to the same data center, the deployment cost and energy consumption can be reduced. In this paper, we propose a two-phase framework to find optimal base station clustering schemes in a city-wide Cloud-RAN. First, we design a traffic profile for each base station, and propose an entropy-based metric to characterize the complementarity among base stations. Second, we build a graph model to represent the complementarity as link weight, and propose a distanceconstrained clustering algorithm to find optimal base station clustering schemes. We evaluate the performance of our framework using two months of real-world mobile network traffic data in Milan, Italy. Results show that our framework effectively reduces 12.88% of deployment cost and 9.45% of energy consumption compared with traditional architectures, and outperforms the baseline method.

No abstract