<ul><li>A deep learning framework called EMForecaster has been developed for time series forecasting in wireless networks.</li><li>EMForecaster employs patching and reversible instance normalization and mixing operations for efficient feature extraction.</li><li>The framework includes a conformal prediction mechanism for uncertainty quantification of forecasts.</li><li>EMForecaster outperforms current state-of-the-art DL approaches and achieves a superior balance between prediction interval width and coverage.</li></ul>

EMForecaster: A Deep Learning Framework for Time Series Forecasting in Wireless Networks with Distribution-Free Uncertainty Quantification

Discover more