When

31 October 2025, 17:30 Athens time

Where

• Remote via Zoom Link (Meeting ID: 923 5830 3658, Password: 189175)
• MLDS Students: Science Building 145Π58

Abstract

Learn the basics of practical machine learning and deep learning for classification problems in MATLAB. Use a machine learning model that extracts information from real-world data or build and train a deep neural network to group your data into predefined categories.

• Fundamentals of Artificial Intelligence
• Example of using unlabeled data
• Classification example using machine learning
• Classification example using deep learning
• Further resources (MATLAB Copilot, Academic Training Suite)

Bring your laptop along to run the code in real time. 

When

8 November 2024, 17:00 Athens time

Where

• Remote via Zoom Link (Meeting ID: 927 7232 1162, Password: 673118)
• MLDS Students: Science Building 145Π58

Abstract

Controlling large teams of robots is a crucial challenge in robotics, due to the need for solutions that balance efficiency, scalability, and robustness. This talk will delve into recent advancements in learning-based control for multi-robot systems, with a focus on scalable coordination and decentralized decision-making. I will present the latest results we have achieved to efficiently learn distributed control strategies for large teams of robots, leveraging physics-informed machine learning and generative AI techniques. We will see how physics-informed learning can be used to provide the learned controllers three key properties: interpretability, modularity, and scalability. Similarly, we will demonstrate how generative AI can be used to ease interaction with non-expert users to describe desired large-scale swarm configurations, producing smooth trajectories and accounting for potential collisions through a reactive navigation algorithm.

Bio

Eduardo Montijano is an Associate Professor in the Departamento de Informática e Ingeniería de Sistemas at Universidad de Zaragoza in Spain. He received the M.Sc. and Ph.D. degrees from the Universidad de Zaragoza, Spain, in 2008 and 2012 respectively. He was a faculty member at Centro Universitario de la Defensa, Zaragoza, between 2012 and 2016. His research interests are in the field of distributed algorithms applied to cooperative control and perception of multi-robot systems.

When

11 October 2024, 17:00 Athens time, Science Building 145Π58

Abstract

Information Retrieval (IR) is a part of data science that involves the process of obtaining relevant information from large collections of unstructured or semi-structured data, such as documents, web pages, or multimedia, based on a user’s query, typically using algorithms for indexing, searching, ranking, and filtering data to improve search relevance and efficiency. In this seminar, we will make an introduction to the basic concepts of IR and build a basic IR system, without coding, using Altair RapidMiner Studio. For our case study, we will use a medical text base, namely the CFC Cystic Fibrosis text base. CFC Cystic Fibrosis is a specialized collection of scientific and medical abstracts, bibliographic records, and research articles focused on cystic fibrosis, covering topics such as clinical studies, microbiology (e.g., Pseudomonas aeruginosa), treatment outcomes, and related biomedical research. Participants will be engaged in a hands-on experience following up the steps of IR system development on par with the lecturer. Therefore, they are encouraged to have downloaded and installed RapidMiner Studio and its Text Processing Extension.

Short Bio

Nikos Giatrakos is an Assistant Professor at the School of Electrical and Computer Engineering of the Technical University of Crete (Greece). He received his BSc Degree in Computer Science from the University of Piraeus (Greece) in 2006, the MSc degree in Information Systems from the Athens University of Economics and Business (Greece) in 2007, and the PhD degree in Computer Science from the University of Piraeus (Greece) in 2012. His research interests are in the broad area of Big Data Management algorithms, software architectures and systems including Big streaming Data & Real-Time Analytics, Distributed/Decentralized Big Data Processing, Federated Machine Learning, Edge-to-Cloud Big Data Management, Synopses for Massive Data/Approximate Query Processing, Complex Event Processing. He was a recipient of the Best Demo Award in ACM CIKM 2020. He has contributed as one of key investigators in several recent EU projects, he has served as the co-coordinator of the EU H2020 project INFORE and as a Principal Investigator at the EU Horizon project EVENFLOW.

1. "Numerical and Lyapunov-based investigation of the effect of stenosis on blood transport stability using a control-theoretic PDE model of cardiovascular Flow" by Dr. Shantanu Singh

Bio

Shantanu Singh is post-doc in Department of Electrical and Computer Engineering at Technical University of Crete. He completed his PhD from Tel Aviv University, Israel. He was a Marie-Curie fellow from 2018 to 2022. He received his master’s degree in Electrical Engineering at Indian Institute of Technology, New Delhi. His research interests are distributed parameter systems with applications to boundary control systems, flow systems and hyperbolic partial differential equations. 

Title

Numerical and Lyapunov-based investigation of the effect of stenosis on blood transport stability using a control-theoretic PDE model of cardiovascular Flow

Abstract

We perform various numerical tests to study the effect of (boundary) stenosis on blood flow stability, employing a detailed and accurate, second order finite-volume scheme for numerically implementing a partial differential equation (PDE) model, using clinically realistic values for the artery's parameters and the blood inflow. The model consists of a baseline 2x2 hetero-directional, nonlinear hyperbolic PDE system, in which, the stenosis' effect is described by a pressure drop at the outlet of an arterial segment considered. We then study the stability properties (observed in our numerical tests) of a reference trajectory, corresponding to a given time-varying inflow (e.g., a periodic trajectory with period equal to the time interval between two consecutive heartbeats) and stenosis severity, deriving the respective linearized system and constructing a Lyapunov functional. Due to the fact that the linearized system is time varying, with time-varying parameters depending on the reference trajectories themselves (that, in turn, depend in an implicit manner on the stenosis degree), which cannot be derived analytically, we verify the Lyapunov-based stability conditions obtained, numerically. Both the numerical tests and the Lyapunov-based stability analysis show that a reference trajectory is asymptotically stable with a decay rate that decreases as the stenosis severity deteriorates.

2. "Simultaneous Compensation of Input Delay and State Quantization for Linear Systems via Switched Predictor Feedback" by Dr. Florent Koudohode

Bio

Florent Koudohode received his Ph.D. degree in Automatic Control from LAAS-CNRS, Université Paul Sabatier, Toulouse, France, in 2023. He completed two M.Sc. degrees in fundamental mathematics and applications, one from IMSP, Dangbo, Benin in 2018, and the other from Université Paul Sabatier, Toulouse III, France in 2020. He is currently a Postdoctoral Researcher in the Department of Electrical and Computer Engineering at the Technical University of Crete, Greece. His research interests include distributed parameter systems, event-triggered control of PDE, quantization, time-delays systems and nonlinear control.

Title

Simultaneous Compensation of Input Delay and State Quantization for Linear Systems via Switched Predictor Feedback

Abstract

We develop a switched predictor-feedback law, which achieves global asymptotic stabilization of linear systems with input delay and with the plant and actuator states available only in (almost) quantized form. The control design relies on a quantized version of the nominal predictor-feedback law for linear systems, in which quantized measurements of the plant and actuator states enter the predictor state formula. A switching strategy is constructed to dynamically adjust the tunable parameter of the quantizer (in a piecewise constant manner), in order to initially increase the range and subsequently decrease the error of the quantizers. The key element in the proof of global asymptotic stability in the supremum norm of the actuator state is derivation of solutions' estimates combining a backstepping transformation with small-gain and input-to-state stability arguments, for addressing the error due to quantization. 

When: Friday, 04.10.2024.

Time: 17:00 pm

Location: Room 145Π58

On Friday 27 September (at 17.15 pm, in Room 145Π58), Professor Dionysios Christopoulos, Director of the MSc in Machine Learning and Data Science will meet the MLDS students.

European Association for International Education (EAIE) is the European center for expertise, networking and resources in the internationalization of higher education. 

Study in Greece (SiG) is the organization of Greek higher education institutions, dedicated to promoting internationalization of Greek higher education. SiG operates as a non-profit organization comprising representatives from all 24 Greek state universities, and is governed by a dedicated board of directors who bring knowledge and expertise from leadership positions in Greek universities.

For additional information, you may visit the SiG website.

Time: 17:00 pm

Location: Room 145Π58

The list of the presentations includes:

• Argyro Talamagka: Comparative Analysis of Sunspot Number Predictions with Machine Learning Models
• Konstantinos Palaiologos: Clustering Analysis and Predictive Modeling of Global Fisheries Data
• Christoforos Zisis: A framework for Optimal and Efficient Neural Architecture Search
• Thrasyvoulos Giakoumakis: Time Series Analysis and Machine Learning Approaches in Weather Prediction for the area of Perth (Australia) Using Historical Data
• Ioanna Dalamagka: Analysis of time series data from an air pollution network
• Nafiseh Pourmousa : Forecasting and Visualization of Traffic Relationship over City Road Network Time Series

Streaming will be available at: 
https://tuc-gr.zoom.us/j/92645009781?pwd=P0uopr9dUgUIk3c6XUKJdPEnOrw0Cv.1

Meeting ID: 926 4500 9781
Password: 148925

Dr. Nikos Giatrakos
Assistant Professor at the School of Electrical and Computer Engineering, Technical University of Crete and a member of the Software Technology and Network Applications Lab (Soft Net).

When

11 July 2024, 5th ACM Europe Summer School on Data Science, at the Panorama Hall of the Royal Olympic Hotel in Athens

Title

Proactive Streaming Analytics at Scale: A Journey from the State-of-the-art to a Production Platform

Abstract

Proactive streaming analytics continuously extract real-time business value from massive data that stream in data centers or clouds. This requires (a) to process the data while they are still in motion; (b) to scale the processing to multiple machines, often over various, dispersed computer clusters, with diverse Big Data technologies; and (c) to forecast complex business events for proactive decision-making. Combining the necessary facilities for proactive streaming analytics at scale entails: (I) deep knowledge of the relevant state-of-the-art, (II) cherry-picking cutting edge research outcomes based on desired features and with the prospect of building interoperable components, and (III) building components and deploying them into a holistic architecture within a real-world platform. In this tutorial, we drive the audience through the whole journey from (I) to (III), delivering cutting edge research into a commercial analytics platform, for which we provide a hands-on/demo experience.

Applications are submitted on the online platform of the program.

For more information, you may visit the 5th ACM Europe Summer School website.

This PhD project will assess the impacts of climate change on renewable energy resources in Morocco. The candidate will analyze historical climate data, model future scenarios, and evaluate implications for solar, wind, and hydroelectric energy production. The goal is to provide actionable insights for sustainable energy planning and policy-making in Morocco.

2. Defining Suitable Sites for Rainwater Harvesting in Arid Regions

This research aims to identify optimal sites for water harvesting in arid and semi-arid regions. The candidate will use geographic information systems (GIS), remote sensing, and hydrological modeling to locate potential water harvesting sites. This work is crucial for improving water security and supporting agricultural and community needs in water-scarce areas.

3. Uncertainty in Regional Climate Simulation

This PhD position focuses on quantifying and reducing uncertainties in regional climate simulations. The candidate will work with regional climate models (RCMs) to evaluate their performance under different scenarios and develop methods to improve accuracy. This research will contribute to more reliable climate projections, essential for informed decision-making.

4. Use of Stochastic Methods for Data Assimilation to Improve Flood Forecasting

This project will investigate the application of stochastic methods for data assimilation to enhance flood forecasting accuracy. The candidate will explore various techniques, such as ensemble Kalman filtering and particle filtering, to integrate observational data with hydrological models. The objective is to develop robust forecasting tools to better predict flood events and mitigate their impacts.

Application Process:

Interested candidates are invited to submit their applications through the following links:

Subject 1: https://career2.successfactors.eu/sfcareer/jobreqcareer?jobId=11471&company=ump

Subject 2:  https://career2.successfactors.eu/sfcareer/jobreqcareer?jobId=11471&company=ump

Subject 3: https://career2.successfactors.eu/sfcareer/jobreqcareer?jobId=11550&company=ump

Subject 4: https://career2.successfactors.eu/sfcareer/jobreqcareer?jobId=11549&company=ump