Special Session on Practical Applications of Deep Learning (PADL) at IDEAL 2020

PADL@IDEAL 2020


Artificial Intelligence



Deep Learning provides a great deal of current research on Machine Learning and its applications. The possibilities offered by Deep Learning architectures such as Convolutional Neural Networks, Recurrent Neural Networks or Deep Belief Networks have allowed to successfully tackle a plethora of complex problems in the last two decades. Nevertheless, the applications of DL have not stopped growing and new architectures are presented to deal with problems not completely solved yet. Deep Learning has excelled in its extraordinary capacity in tasks such as image classification, audio preprocessing, malware detection, fraud prediction or machine translation, among many others. In the light of the promising results shown in these tasks, sometimes restricted to a test environment, these deep architectures have extended gradually to real scenarios.
In this Special Session, we expect to see research geared towards effective and practical Deep Learning solutions for current and new application domains. We aim to take a step forward on the use of these deep architectures in order to provide useful instruments in domains such as Internet of things (IoT), cyber-physical systems (CPS), cybersecurity, medical diagnosis, fake news detection, or in any other practical domain booming nowadays. On the whole, this Special Session will bring together researchers interested in Deep Learning and its applications, creating an inspiring environment to share ideas and to seek collaborations.
Topics of interest include, but are not limited to:
- Industrial Deep Learning-based Applications for IoT
- Deep learning-based anomaly detection for CPSs
- Process mining supported by Deep Learning
- Deep learning for time series classification and forecasting
- Deep Learning for Energy Consumption Optimisation
- Practical deep learning for natural language processing tasks
- Practical deep learning for computer vision tasks
- Practical deep learning for Multimodal Learning tasks
- Interpretability and Explainability of real-world Deep Learning solutions
Alejandro Martín (Universidad Politécnica de Madrid, Spain) alejandro.martin@upm.es
Víctor Rodriguez-Fernandez (Universidad Politécnica de Madrid, Spain) victor.rodriguezf@uam.es