Jelena PESIC  (Nokia France)

Will Machine Learning mitigate the extra cost of increase in capacity?

Abstract
Machine learning has become an exciting subject in optical networks for both industry and academia. Its success reflects the growing need for network adaptation to changing circumstances (e.g., traffic, topology, or even user preferences). Moreover, challenges in optical networks are becoming more critical as we think about their evolution towards future scenarios with more data traffic, such as 5G networks, IoT, and others. Tomorrow’s optical networks are a complex domain asking for very high capacity networks that also need to save the cost per bit. In this talk, we will try to explore machine learning for optical networks in light of its enormous technical capabilities as well as its potential to reduce costs.