This hands-on course takes you from transformer fundamentals to cutting-edge agentic AI systems. You’ll learn how large language models work under the hood, train and fine-tune your own models, and build autonomous agents that use tools, access external resources, and collaborate to solve complex tasks. Each session combines theory with practical implementation using industry-standard frameworks like Hugging Face, LangChain, and the Model Context Protocol.
Session 1: Foundations & Architecture Understanding Transformers: The Engine Behind Modern AI
Discover how LLMs predict and generate text through self-attention mechanisms. We’ll demystify the transformer architecture—from tokenization to embeddings—and you’ll implement a tokenizer from scratch while visualizing how models “pay attention” to different parts of text.
Session 2: Training Fundamentals Training Your Own Language Model
Learn what it takes to train an LLM: pre-training objectives, dataset curation, and scaling laws. You’ll fine-tune a real model (GPT-2 or TinyLlama) on custom data using Hugging Face tools, understanding the trade-offs between model size, compute, and performance.
Session 3: Advanced Training & Alignment Making Models Helpful: Instruction Tuning & RLHF
Explore how base models are transformed into helpful assistants through instruction tuning and alignment techniques like RLHF. You’ll fine-tune a model to follow instructions and experiment with advanced prompting techniques including chain-of-thought reasoning.
Session 4: Cutting-Edge Applications & Agentic Systems Autonomous AI: Building Agents That Think and Act
Go beyond chatbots to autonomous agents that use tools, access databases, and collaborate on complex tasks. You’ll implement function calling, integrate the Model Context Protocol (MCP) to connect LLMs with external resources, and build multi-agent systems using LangGraph—culminating in an autonomous agent demo.
Lecturer
Kerem Şenel received his PhD in Computer Science from LMU Munich in 2025, specializing in Natural Language Processing. During his doctoral research at the Center for Information and Language Processing (CIS), he explored diverse topics including interpretability, multilinguality, and evaluation of language models. He currently works as an IT consultant in industry, specializing in AI applications and solutions.
This course provides a comprehensive overview of how Virtual Reality (VR) intersect with Brain-Computer Interfaces (BCIs), which are neurotechnologies that introduce promising possibilities to interact with digital devices solely through the acquisition and analysis of brain activity, typically measured using electroencephalography.
BCIs enhance VR applications in two key ways: by enabling direct control of virtual elements through mental commands, such as imagining hand movements to guide a virtual character, and by gathering real-time neural data to adapt and personalize the VR experience to the user\’s cognitive and emotional state. Conversely, VR platforms offer immersive environments that facilitate BCI user training and rehabilitation, creating tailored scenarios that improve brain activity modulation and learning outcomes.
The course is structured into four sessions: the first two cover the mutual benefits of integrating BCI with VR technologies. The final two sessions will focus on therapies that utilize VR-based and BCI-based approaches independently, as well as innovative interventions at the intersection of both technologies. These combined VR-BCI therapies harness neurofeedback, and immersive environments to promote functional recovery, for instance in motor rehabilitation after stroke. This integrated approach provides patient-centered, adaptable, and motivating rehabilitation protocols that leverage real-time brain activity monitoring to enhance neuroplasticity and clinical outcomes.
Session 1 and 2: mutual benefits of integrating BCI with VR technologies Sessions 3 and 4: state of the art on VR and BCI-based therapies
Literature
• Drigas, A., & Sideraki, A. (2024). Brain neuroplasticity leveraging virtual reality and brain–computer interface technologies. Sensors, 24(17), 5725.
• Kober, S. E., Wood, G., & Berger, L. M. (2024). Controlling virtual reality with brain signals: state of the art of using VR-based feedback in neurofeedback applications. Applied psychophysiology and biofeedback, 1-20.
• Lotte, F., Faller, J., Guger, C., Renard, Y., Pfurtscheller, G., Lécuyer, A., & Leeb, R. (2012). Combining BCI with virtual reality: towards new applications and improved BCI. In Towards practical brain-computer interfaces: Bridging the gap from research to real-world applications (pp. 197-220). Berlin, Heidelberg: Springer Berlin Heidelberg.
• Roc, A., Pillette, L., Mladenovic, J., Benaroch, C., N’Kaoua, B., Jeunet, C., & Lotte, F. (2021). A review of user training methods in brain computer interfaces based on mental tasks. Journal of Neural Engineering, 18(1), 011002.
Lecturer
Dr. Léa Pillette is a CNRS researcher and member of the Seamless team at IRISA, Rennes, France, since 2022. She obtained her PhD in computer science from the University of Bordeaux in 2019. Her research focuses on developing innovative methods to train individuals to regulate their brain activity, enabling more accessible and effective use of brain-computer interfaces for applications such as medical interventions and virtual world interactions.
Lecturer: Klaus Gramann Fields: Cognitive Neuroscience, Mobile Brain Imaging
Content
The human brain has evolved to optimize the outcome of our behavior. Yet, established human brain imaging approaches restrict any active movement of participants to avoid artifacts from distorting the signal of interest. Recent developments in brain imaging technologies allow for conducting experiments beyond established laboratory-based experimental protocols. Light-weight mobile EEG and fNIRS amplifiers can be combined with additional modalities like motion capture, eye tracking and virtual reality providing unprecedented insights into behavioural and brain dynamic states during embodied interactions with our surroundings. The course will introduce Mobile Brain/Body Imaging (MoBI). The core knowledge and skills taught by the course are: • the fundamental concepts behind EEG and problems related to movement • the basic concepts of MoBI (embodiment, technology, applications) • core ideas and findings in MoBI research • application of MoBI to the field of spatial cognition
In more detail, the course will have four sessions with the following topics: 1. Fundamental EEG Concepts: physiological origins of the EEG signal, generators, volume and capacitive conduction, oscillations and origins, extraction of time domain and frequency domain parameters 2. Fundamental EEG Concepts: EEG technology, traditional amplifiers, newer developments 3. Basic concepts of MoBI: Embodiment, EEG and movement, multimodal Data acquisition, multimodal data analyses 4. MoBI application: Embodied spatial navigation, MoBI and traditional desktop comparison, newer developments
Literature
Literature is optional and more regarded as ‘further/complementary reading’:
1) Wilson, M. (2002). Six views of embodied cognition. Psychonomic bulletin & review, 9(4), 625-636.
2) Niso, G., Romero, E., Moreau, J. T., Araujo, A., & Krol, L. R. (2023). Wireless EEG: A survey of systems and studies. NeuroImage, 269, 119774.
3) Makeig, S., Gramann, K., Jung, T.-P., Sejnowski, T.J., & Poizner, H. (2009). Linking Brain, Mind and Behavior. International Journal of Psychophysiology, 73(2), 95-100.
4) Gramann, K. (2024). Mobile EEG for Neurourbanism Research-What Could Possibly Go Wrong? A Critical Review with Guidelines. Journal of Environmental Psychology, 102308.
Lecturer
Klaus Gramann
seit 07/2012 Professor of Biological Psychology and Neuroergonomics, Technical University Berlin, Germany 10/2011 – 04/2012 Acting Professor of Cognitive Psychology, University Osnabrück, Germany 05/2011 – 10/2011 Visiting Professor, National Chiao Tung University, Hsinchu, Taiwan 07/2011 – 12/2011 Associate Research Scientist, University of California, San Diego, USA 05/2007 – 07/2011 Assistant Research Scientist, University of California, San Diego, USA 03/2004 – 05/2007 Assistant Professor (C1), University Munich, Germany 06/2002 – 03/2004 Post Doctoral Scholar, University Munich, Germany 2002 – 2007 Habilitation Biological and General Psychology, University Munich, Germany 1998 – 2002 Ph.D. Psychology, Technical University Aachen, Germany 1998 – 1998 Diploma Psychology, Justus Liebig University Gießen, Germany 1994 – 1996 Pre-Diploma Psychology, Justus Liebig University Gießen, Germany 1991 – 1993 Certified Communication Manager, Academy for Communication, Kassel, Germany
The course Introduction to Intelligent User Interfaces (IUI) introduces participants to key concepts at the intersection of Human-Computer Interaction (HCI) and Artificial Intelligence (AI). It explores how methods from Machine Learning and AI can be transferred to the design of interactive systems that act intelligently, adapt to users, and support human goals. Emphasis is placed on a human-centered perspective that prioritizes usability, transparency, and user trust. Across four sessions, participants will gain a conceptual understanding of the foundations, design principles, and open challenges of intelligent user interfaces, preparing them to critically assess and discuss current and future developments in this field. * 1. Session: Motivation and Introduction * 2. Session: Machine Learning and Human-Computer Interaction basics * 3. Session: Designing, Building, and Evaluating Human-AI Systems * 4. Session: Human-Centered Challenges and Future Directions
Literature
Andy Field and Graham Hole (2002). How to Design and Report Experiments
Kasper Hornbæk, Per-Ola Kristensson, and Antti Oulasvirta (2025). Introduction to Human-Computer Interaction. Oxford University Press.
Sven Mayer is a full professor of computer science at the TU Dortmund University (Germany) and the Research Center Trustworthy Data Science and Security, where he is the head of the chair for Human-AI Interaction. His research focuses on Human-AI Interaction at the intersection between Human-Computer Interaction and Artificial Intelligence, where he focuses on the next generation of computing systems. He uses artificial intelligence to design, build, and evaluate future human-centered interfaces. In particular, he envisions enabling humans to outperform their performance in collaboration with the machine. He focuses on areas such as augmented and virtual reality, mobile scenarios, and robotics.
Matthias Schmidmaier is a research engineer in the Human-AI Interaction group of Prof. Sven Mayer at the Research Center Trustworthy Data Science and Security, TU Dortmund University, with a research affiliation at LMU Munich. His current work explores empathic interaction with intelligent systems, such as mental health chatbots or social robots, investigating the need, impact, and sources of perceived empathy and related constructs. Alongside his academic research, he brings over a decade of industry experience developing human-centered technologies in affective computing, vision-based behavior analytics, and XR, across both dynamic startup environments and collaborations with major international companies.
Where do apparently opposite qualities of experience show up in our inner lives? How do thinking and feeling fit together — and where don’t they? What shapes the connection between body and mind? How do we experience the outer world, and what do we experience within — and how can these be linked and integrated? What stays unconscious, and what becomes conscious? These and similar questions are at the heart of our self-experience workshop. Based on experiential exercises drawing from psychoanalysis, humanistic psychology, and body-oriented approaches, participants are invited into a reflective space to explore self-awareness, perception, and communication. No prior knowledge is required — all you need is a little curiosity and a willingness to gently step beyond the edge of your comfort zone. Then self-experience can become a bridge to new realities of being and relating.
Lecturer
Katharina Krämer is a psychologist and analytic psychotherapist. She works as a professor for psychology at the Rheinische Hochschule Köln, University of Applied Sciences, Cologne, Germany, and as a lecturer and supervisor for psychotherapists in training. Additionally, she works as a psychotherapist in private practice. In 2014, Katharina Krämer received her doctoral degree from the University of Cologne, Germany, on a thesis investigating the perception of dynamic nonverbal cues in cross-cultural psychology and high-functioning autism. Her research interests include the application of Mentalization-Based Group-Therapy with patients with autism and the vocational integration of patients with autism.
Annekatrin Vetter is a clinical psychologist and analytic psychotherapist. As a psychotherapist, she treats patients with different mental disorders in private practice. Additionally, she works as a lecturer and supervisor for psychotherapists in training and as a trainer for Coaches at Inscape – Coaching & Counselling, Cologne, Germany.
Affiliation: Praxis für Psychotherapie und Psychoanalyse, Supervision und Coaching Annekatrin Vetter, Cologne
Sophia Reul is a clinical psychologist and analytic psychotherapist. She works as a psychotherapist in private practice. In 2021, she received her doctoral degree from the Westfälische Wilhelms-University Münster, Germany, on a thesis investigating the impact of neuropsychological methods in diagnoses of early dementia. Today, her research interests include the application of Mentalization-Based Group-Therapy (MBT-G) with patients with autism.
Affiliation: Praxis für Psychotherapie und Psychoanalyse Sophia Reul, Kirchweidach (Bay.)
Lecturer: Terrence Stewart Fields: Computational Neuroscience, Neuroscience, AI
Content
This course provides an overview of computational neuroscience, the science of creating computer simulations of neurons, groups of neurons, and different brain systems, and then comparing the results of these simulations to the behaviour of real brains. This lets us better understand how brains work, and it also has the potential of inspiring new types of Artificial Intelligence systems.
We start by looking at individual neurons and their details, then move to the three major approaches to making large-scale models capable of producing detailed behaviour: Parallel Distributed Processing (PDP++/Emergent), Dynamic Neural Fields (DNF/Cedar), and the Neural Engineering Framework (NEF/Nengo). Python notebooks will be provided for hands-on examples.
Session 1: Individual neurons Session 2: Many neurons in parallel (PDP++) Session 3: Dynamic Neural Fields (DNF) Session 4: The Neural Engineering Framework and Nengo
Literature
Kriegeskorte, N., & Douglas, P. K. (2018). Cognitive computational neuroscience. Nature neuroscience, 21(9), 1148–1160. https://doi.org/10.1038/s41593-018-0210-5
Rumelhart, D., & McClelland, J., (1986). Parallel distributed processing: Explorations in the microstructure of cognition. MIT Press, Cambridge, MA, USA
Schöner, G. (2023). Dynamical Systems Approaches to Cognition. In Sun, Ron (Ed.), The Cambridge Handbook of Computational Cognitive Sciences (2nd ed.). Cambridge University Press.
Stewart, T.C., & Eliasmith, C. (2014). Large-scale synthesis of functional spiking neural circuits. Proceedings of the IEEE, 102(5):881–898.
Lecturer
Terry Stewart is a Senior Research Officer at the National Research Council Canada, and Site Lead of the NRC-University Waterloo Collaboration Centre. His research includes large-scale brain simulation, cognitive modelling, energy-efficient neuromorphic computing, and AI safety.
Lecturer: Mike Ambinder Fields: BCI, Video Games, Neuroscience, AI, ML, Engineering
Content
This course will cover one possible future for video game play. At the moment, video games provide a dynamic experience on several axes, but they remain relatively static with respect to the individual player experience. They are designed for the collective – they do not adapt. With the advent of improved hardware, statistical techniques, and advances in game design, the potential exists to design a new generation of gameplay where the experience is tailored to the individual as a consequence of physiological measurement of internal state. Under this framework; games may become capable of a whole lot more than entertainment.
Dr. Mike Ambinder received a BA in Computer Science and Psychology from Yale University and a PhD in Psychology (Visual Cognition) from The University of Illinois. He spent 15 years at Valve leading research efforts in applied psychology and game design, statistics, machine learning, and AI, economic systems design, and Brain-Computer Interfaces. He is currently the Chief Research Officer of Cognitive Explorations, a design consultancy in the games space, and the Chief Research Officer of August Interactive, a gaming startup with a focus on prosocial behavior change.
Affiliation: Cognitive Explorations, LLC; University of Washington; August Interactive
Lecturer: Timothy Drysdale Fields: Artificial intelligence, education, practical work
Content
This evening talk will reflect on the challenge facing educators, particularly younger educators with many years of teaching ahead of them. The joint pressure of readily-available artificial intelligence affecting the validity of traditional processes, and massification of education reducing the resources available per student, pose a difficult pinch point that is generating demand for authentic, interactive activities but placing a lot of pressure on the available time and space for students to experiment with real equipment in a traditional manner. I\’ll introduce a solution in the form of laboratories in a box, which we have been doing doing at the University of Edinburgh for a number of years, and describe the elements that make these successful for us, how you can adopt a similar approach, the pitfalls to avoid and some fruitful future directions for our communities of educators to explore, in particular in expanding what we do with the data streams to support better learning and in taking our concept of experiments beyond what we are used to doing in traditional laboratories.
Literature
Reid, D., & Drysdale, T. (2024). Student-facing learning analytics dashboard for remote lab practical work. IEEE Transactions on Learning Technologies, 17, 1037-1050. https://doi.org/10.1109/TLT.2024.3354128
D.Reid, J. Burridge, D. Lowe, T. Drysdale “Open-source remote laboratory experiments for controls engineering education,” International Journal of Mechanical Engineering Education. February 2022. doi:10.1177/03064190221081451
T. D. Drysdale, S. Kelley, A.-M. Scott, V. Dishon, A. Weightman, R. J. Lewis & S. Watts “Opinion piece: non-traditional practical work for traditional campuses,” Higher Education Pedagogies, 5:1, 210-222, 2020, DOI: 10.1080/23752696.2020.1816845
G. L. Knight & T. D. Drysdale The future of higher education (HE) hangs on innovating our assessment – but are we ready, willing and able?, Higher Education Pedagogies, 5:1, 57-60, 2020, DOI: 10.1080/23752696.2020.1771610
Lecturer
Prof Timothy Drysdale is the Chair of Technology Enhanced Science Education and Director of Strategic Digital Education in the School of Engineering. His main research activity is in Engineering Education, where he leads the Remote Laboratories group. He and his team have developed an entirely new infrastructure and approach for operating online remote laboratories on traditional campuses (practable.io), winning international awards from the Global Online Laboratories Consortium (Remote Experiment Award 2024) and the Association for Learning Technology / Jisc Award for Digital Transformation in 2023. His prior research activities were in the area of terahertz component design and testing, microwave antennas, and optical plasmonics. He has a long-standing involvement with public outreach in science and engineering, including the Royal Society Summer Science Exhibition, Science Day at Buckingham Palace, and giving the Isambard Kingdom Brunel Award Lecture at the British Science Festival.
Lecturer: Jutta Kretzberg, Katja Hellekes Fields: Personal / professional development
Content
Are you a student? Have you ever considered doing a PhD? Or a career in academia? Does the idea of doing a PhD appeal to you? Or does it seem like hard work, or even a painful experience? Many Master’s students struggle with the decision of whether a PhD would be the right choice for their career. In fact, a significant proportion of PhD students continue to question their decision until they graduate, and sometimes even afterwards. There is no general advice on who should pursue a PhD. Whether to pursue a PhD is a personal decision that depends on factors such as your personality, personal situation, and the job opportunities available. The aim of this workshop is to help you develop a clearer personal perspective on this decision.
This workshop is primarily aimed at Master’s and advanced Bachelor’s students. However, the method of developing your personal perspective can also be applied to future career steps. PhD students, PhD holders and non-PhDs who are willing to share their perspectives are highly welcome!
Session 1: Background information In the first session, we will begin by providing some background information on undertaking a PhD in Germany or Austria. What are the motivations for pursuing a PhD? What skills are gained through a PhD? How can a PhD be structured and funded? How do PhDs differ between disciplines and countries?
Session 2: External perspectives In the second session, we will explore the different stakeholders’ perspectives interactively. What do Master’s students expect from a PhD? What do PhD supervisors expect from their students? What do employers expect from PhD versus Master’s degree applicants? What about the perspective of family and friends? And which personality traits might be useful for pursuing a PhD?
Session 3: Your personal perspective During the third session, you will write down your hopes, neutral expectations and fears relating to a PhD. Working with a fellow participant, categorize these into the groups: ‘tasks/skills’, ‘topics/scientific questions’, ‘working environment’ and ‘personal factors’. Sharing your thoughts and listening to those of your teammate can help you gain a clearer perspective on your career decisions.
Session 4: How to become a PhD candidate? After sharing our conclusions from the previous sessions, we will discuss the practical steps involved in becoming a PhD candidate, such as: How do you choose a topic? How do you find a project and a supervisor? How can you finance the PhD? We will also consider how to balance the demands of your PhD with your personal life – bridging realities of your live and a PhD.
Literature
The European Competence Framework for Researchers: https://research-and-innovation.ec.europa.eu/document/download/7da29338-37bf-4d51-b5eb-a1571b84c7ad_en?filename=ec_rtd_research-competence-presentation.pdf
General information on PhD scholarships (by German Government): https://www.bmbf.de/EN/Research/ScienceSystem/AcademicCareers/DoctoralScholarships/doctoralscholarships_node.html
General information on German academic system & funding for international exchange (DAAD): https://www.daad.de/en/
Largest scholarship organisation in Germany: https://www.studienstiftung.de/en/doctoral-scholarships/doctoral-scholarships
Chris Woolston: “Graduate survey: A love-hurt relationship” Nature 550, 549-552 (2017)
https://www.nature.com/articles/nj7677-549a https://doi.org/10.1038/nj7677-549a (Nature’s survey of more than 5,700 doctoral students worldwide)
Lars Kiewidt, PhD: “To PhD or not to PhD?” (2019) https://medium.com/age-of-awareness/to-phd-or-not-to-phd-4312cdb862c5 (Evaluation of this survey data set concerning PhD student’s motivation, skills and satisfaction across fields in natural sciences.)
Chris Woolston: ‘I don’t want this kind of life’: graduate students question career options
Nature 611, 413-416 (2022) doi: https://doi.org/10.1038/d41586-022-03586-8 (Newest version of nature’s PhD survey, but not open access)
Katie Mitzelfelt, PhD: “To Be or Not To Be a PhD Candidate, That Is the Question” (Association for Women in Science Magazine, 2021): https://awis.org/to-phd-or-not-phd/ (Individual perspectives of 6 persons on their own decision to be or not to be a PhD.)
Charlotte King_: “To PhD or not to PhD, that is the question…” https://www.postgrad.com/blog/to-phd-or-not-to-phd/ (Rather old, but still helpful blog post)
Lecturer
Jutta Kretzberg is professor for Computational Neuroscience and head of the MSc program Neuroscience at University of Oldenburg. She studied applied computer science and biology at University of Bielefeld, where she also did her PhD in Biology. After being a postdoc (and having a baby) in San Diego, California, she came back to Germany to be a junior professor and became a professor some years (and another baby) later. Nowadays, while juggling her family, teaching, research and administration duties, her favorite task is mentoring.
Katja Hellekes, is an experienced academic professional and Coordinator of the Vienna Doctoral School Cognition, Behavior and Neuroscience. She completed her diploma and doctorate at the University of Cologne, specializing in Neurobiology, followed by postdoctoral research at the Institute of Molecular Pathology (IMP) and the University of Freiburg. Alongside her role as Coordinator of the Doctoral Program, Katja Hellekes lectures in cognitive science. With a passion for fostering the growth of early-career researchers, she provides dedicated support to PhD candidates, guiding them through their doctoral journey and helping them transition into independent research roles.
Lecturer: Jennifer Fewell Fields: Biology; Collective Behavior
Content
This course will explore the social organization and collective behavior of social insects from a biological perspective. The social insects are models for coordination and cooperation across small to large scales. Their distributed communication systems have been used extensively as inspiration for applied questions in coordination and collective behavior, from supply chains to robotics and beyond. Is a social insect colony the original collective \”AI\”? – well probably not, but it will be a fun question to explore!
Lecturer
Jennifer Fewell is a President\’s Professor at Arizona State University, where she served as the founding Director of the Center for Social Dynamics and Complexity. She has served also as President of the Animal Behavior Society and the International Union for the Study of Social Insects. She studies social organization and division of labor in social insects, and mechanisms and evolution of social cooperation across a range of species. She received her MS and PhD from the University of Colorado.
