Congrats to the three extremely talented students of IMU whose diploma theses were successfully examined today:

• Melina Mai with the thesis “Deep Learning for Business Process Monitoring and Prediction”, which proposed deep learning techniques and specifically recurrent neural networks with Long-Short-Term Memory (LSTM) architecture to predict the next event in a business process and applied it on a real dataset of the banking sector,
• Afroditi Fouka, with the thesis “Real-time failure detection in Industry 4.0 with Deep Neural Networks”, which proposed an approach for real-time prediction of the equipment health state using time-domain features extraction, LSTM neural networks, and Bayesian Online Changepoint Detection and applied it to a real-life case in the steel industry and
• Mattheos Fikardos with the thesis “Modelling Decision-Making Processes with Deep Reinforcement Learning”, which used reinforcement learning algorithms (DQN) where an agent was trained to solve decision-making problems and carried out many experiments in order to simulate different application scenarios.

And of course, congrats to Katerina Lepenioti and Alexandros Bousdekis for their continuous supervision and support.