Residual-based Adaptive Huber Loss (RAHL) - Design of an improved Huber loss for CQI prediction in 5G networks
Resumo
The Channel Quality Indicator (CQI) plays a pivotal role in 5G networks, optimizing infrastructure dynamically to ensure high Quality of Service (QoS). Recent research has focused on improving CQI estimation in 5G networks using machine learning. In this field, the selection of the proper loss function is critical for training an accurate model. Two commonly used loss functions are Mean Squared Error (MSE) and Mean Absolute Error (MAE). Roughly speaking, MSE put more weight on outliers, MAE on the majority. Here, we argue that the Huber loss function is more suitable for CQI prediction, since it combines the benefits of both MSE and MAE. To achieve this, the Huber loss transitions smoothly between MSE and MAE, controlled by a user-defined hyperparameter called delta. However, finding the right balance between sensitivity to small errors (MAE) and robustness to outliers (MSE) by manually choosing the optimal delta is challenging. To address this issue, we propose a novel loss function, named Residual-based Adaptive Huber Loss (RAHL). In RAHL, a learnable residual is added to the delta, enabling the model to adapt based on the distribution of errors in the data. Our approach effectively balances model robustness against outliers while preserving inlier data precision. The widely recognized Long Short-Term Memory (LSTM) model is employed in conjunction with RAHL, showcasing significantly improved results compared to the aforementioned loss functions. The obtained results affirm the superiority of RAHL, offering a promising avenue for enhanced CQI prediction in 5G networks.
Referências
Cavazza, J. and Murino, V. (2016). Active regression with adaptive huber loss. arXiv preprint arXiv:1606.01568.
Cesa-Bianchi, N. and Lugosi, G. (2006). Prediction, learning, and games. Cambridge university press.
Clevert, D., Unterthiner, T., and Hochreiter, S. (2016). Fast and accurate deep network learning by exponential linear units (elus). In 4th International Conference on Learning Representations (ICLR’16).
De Myttenaere, A., Golden, B., Le Grand, B., and Rossi, F. (2015). Using the mean absolute percentage error for regression models. In ESANN.
Delibalta, I., Gokcesu, K., Simsek, M., Baruh, L., and Kozat, S. S. (2016). Online anomaly detection with nested trees. IEEE Signal Processing Letters, 23(12):1867–1871.
Dong, H. and Yang, L. (2020). Training robust support vector regression machines for more general noise. Journal of Intelligent & Fuzzy Systems, 39(3):2881–2892.
Gokcesu, K. and Gokcesu, H. (2021). Generalized huber loss for robust learning and its efficient minimization for a robust statistics. arXiv preprint arXiv:2108.12627.
Gokcesu, K., Neyshabouri, M. M., Gokcesu, H., and Kozat, S. S. (2018). Sequential outlier detection based on incremental decision trees. IEEE Transactions on Signal Processing, 67(4):993–1005.
Hastie, T., Tibshirani, R., and Wainwright, M. (2015). Statistical learning with sparsity: the lasso and generalizations. CRC press.
Huber, P. J. (2004). Robust statistics, volume 523. John Wiley & Sons.
Intrig-unicamp (2023). Hackathon data collection on 5g network usage in brazil. Available at GitHub: .
Khan, F., Wang, H., and Yang, M. (2016). Application of loss functions in process economic risk assessment. Chemical Engineering Research and Design, 111:371–386.
Kimura, B. Y. L., Almeida, J., et al. (2021). Deep learning in beyond 5g networks with image-based time-series representation. arXiv preprint arXiv:2104.08584.
Meyer, G. P. (2021). An alternative probabilistic interpretation of the huber loss. In Proceedings of the ieee/cvf conference on computer vision and pattern recognition, pages 5261–5269.
Mustafa, R. (2023). Benchmarking youtube in 4g vs 5g through open datasets. Available at GitHub: .
Parera, C., Redondi, A. E., Cesana, M., Liao, Q., and Malanchini, I. (2019). Transfer learning for channel quality prediction. In 2019 IEEE International Symposium on Measurements & Networking (M&N), pages 1–6. IEEE.
Portnoy, S. and He, X. (2000). A robust journey in the new millennium. Journal of the American Statistical Association, 95(452):1331–1335.
Raca, D., Leahy, D., Sreenan, C. J., and Quinlan, J. J. (2020, Available: DOI: 10.1145/3339825.3394938. Beyond throughput, the next generation: a 5g dataset with channel and context metrics. In Proceedings of the 11th ACM multimedia systems conference, pages 303–308.
Raca, D., Zahran, A. H., Sreenan, C. J., Sinha, R. K., Halepovic, E., Jana, R., and Gopalakrishnan, V. (2020). On leveraging machine and deep learning for throughput prediction in cellular networks: Design, performance, and challenges. IEEE Communications Magazine, 58(3):11–17.
Sakib, S., Tazrin, T., Fouda, M. M., Fadlullah, Z. M., and Nasser, N. (2020). A deep learning method for predictive channel assignment in beyond 5g networks. IEEE Network, 35(1):266–272.
Sun, Q., Zhou, W.-X., and Fan, J. (2020). Adaptive huber regression. Journal of the American Statistical Association, 115(529):254–265.
Vankayala, S. K. and Shenoy, K. G. (2020). A neural network for estimating cqi in 5g communication systems. In 2020 IEEE Wireless Communications and Networking Conference Workshops (WCNCW), pages 1–5. IEEE.
Yin, H., Guo, X., Liu, P., Hei, X., and Gao, Y. (2020). Predicting channel quality indicators for 5g downlink scheduling in a deep learning approach. arXiv preprint arXiv:2008.01000.
