Seminar Assignment Summer 2022
The central registration for all computer science seminars will open on March 14th.
This system is used to distribute students among the available seminars. To register for any of the seminars that are offered by the computer science department, you have to register here. You can then select which seminar you would like to take until April 12th 23:59 CET, and will then be automatically assigned to one of them on April 15th.
Please note the following:
- We aim to provide a fair mapping that respects your wishes, but at the same time also respects the preferences of your fellow students.
- Experience has shown that particular seminars are more popular than others, yet these seminars cannot fit all students. We thus encourage students to select their preferences for all available seminars, which eases the process to assign students that do not fit the overly popular seminars to another, less crowded one. Each student must therefore select at least three seminars (with priority from "High" to "Low").
- If you urgently need to be assigned to a seminar in the upcoming semester, choose at least five seminars (with priority from "High" to "Low"). We will then guarantee that you will be assigned to a seminar (yet not necessarily one of your choice).
- If you are really dedicated to one particular seminar, and you do not want any other seminar, please select the "No seminar" as second and third positive option. However, this may ultimately lead to the situation that you are not assigned to any seminar. Also, choosing "No seminar" as second/third option does not increase your chances of getting your first choice.
The assignment will be automatically performed by a constraint solver. You will be added to the respective seminars automatically and be notified about this shortly thereafter. Please note that the assignment cannot be optimal for all students if you drop the assigned seminar, i.e., make only serious choices to avoid penalty to others.
Seminars
Advanced Topics in Program Analysis by Rayna Dimitrova
How can a software developer ensure that the software they have designed does what it is supposed to do? Program analysis is an area of computer science that is concerned with the development of methods and tools that assist programmers in developing correct and robust programs. This includes obtaining a formal understanding of complex programs, automatically verifying that programs work correctly as intended, automatically repairing, generating and optimizing code. Program analysis techniques are nowadays part of the software design process at companies such as Amazon, Facebook, Google, and Microsoft. In this seminar, we will read and discuss research papers that present the state of the art in program analysis.
The seminar will be held in English.
More information: https://cms.cispa.saarland/adv_pa_22/
Requirements: There are no formal prerequisites for this seminar. However, participants are expected to have strong interest in formal verification and logical reasoning, and, having taken the Verification lecture would deffinitely be helpful.
Places: 16
Algorithms for Metagenomics by Vu Lam Dang and Sven Rahmann
This seminar will discuss recent algorithms for metagenomic and metatranscriptomic sequence analysis, as published in conferences like RECOMB, ISMB or WABI. Metagenomics involves finding out the species composition of an environmental sample from DNA sequences alone. Metatranscriptomics refers to examining also the active gene (families) in the mixture of organisms, i.e., to functionally characterize what the community is doing.
The original papers are taken as the basis, but the goal of the seminar is that you work out the core of the method by going beyond the provided paper and do a self-contained presentation of the method. The seminar is intended to be quite technical, so you will need to go into the details of the algorithms and data structures.
Requirements: It will be very helpful to have passed the "Algorithms for Sequence Analysis" lecture before taking this seminar, although it is possible without. In that case, please state your motivation for taking this seminar. Presentation and writing skills are also required.
Places: 12
Algorithms for Realtime Raytracing by Alexander Rath, Ömercan Yazici
With the rise of hardware raytracing new algorithms were created for games and other visualizations with a limited time to render the geometry onto the screen. Waiting hours for a single frame is not an option. Even while modern hardware raytracing helps to get the underlying intersection of the scene fast, a clever algorithm is key to balance interactivity and quality.
In this seminar you will implement, adapt, and evaluate a state-of-the-art algorithm presented in a recent paper. The implementation will be within the AnyDSL framework in the hybrid renderer “Ignis”, which is capable of rendering for hours like offline renderers or for some milliseconds as used in games on a GPU or CPU.
Topics include:
- Sampling (Blue Noise, Anti-Aliasing, ...)
- Light Sampling (ReSTIR, Lightcuts, …)
- Integrators (Surfels, Radiance Caching, ReSTIR GI, …)
- Geometry (Displacement, Hair, …)
- Materials/Media (Subsurface Scattering, Volumes, ...)
- Lens Effects (Polynomial Approximations, …)
- Screen-Space (Stylization, Line-Rendering, ...)
- ...
Together with your implementation you will prepare a demo and a presentation, displaying your achievements to all the participants of the seminar.
Seminar website: https://graphics.cg.uni-saarland.de/courses/rtrt-2022/index.html
Requirements: A strong understanding of computer graphics and good programming skills are essential.
Places: 6
Aspects of Quantitative Program Verification by Benjamin Kaminski
Are you passionate about logic, verification, semantics, and alike? Are you tired of thinking black-and-white, true-or-false? Then come and study the more nuanced quantitative formal program verification! In quantitative verification, properties are not just true or false. Instead, we verify quantities like runtimes, error probabilities, beliefs, etc.
Topics which we will cover include:
- Probabilistic programming (a currently trending modeling paradigm in machine learning),
- The geometry of neural networks
- Incorrectness logic (the latest creation of the former chief formal methods researcher at Facebook)
- The flow of quantitative information through programs
- Worst-case execution times
- Verification of heap-manipulating programs
- and many more
The seminar website can be found here:
Requirements: The following courses are mandatory and/or recommended.
Mandatory: Programmierung 1, Programmiergung 2
Recommended: Grundzüge der Theoretischen Informatik
Nice-to-have: Semantics; Introduction to Computational Logic; Automata, Games and Verification;
Ideal: Verification
Places: 12
Brain Computer Interaction by Maurice Rekrut, Mansi Sharma, Matthias Nadig, Guillermo Reyes Fuentes
Brain-Computer Interfaces (BCI) have been a widely researched topic for the last decades. BCIs make use of brain activity to create controls for computers or machines in general. The most common approach to establish such a connection between computer and brain is to measure the electrical activity of the brain at the scalp surface, via Electroencephalography (EEG). Those EEG devices are non-invasive, comparably cheap and can nowadays even be used with dry electrode caps, which makes them easy to setup. Within this seminar you will learn about different techniques and types of BCIs and work on EEG data that was recorded during a BCI task.
The seminar consists of 3 stages:
In the first stage you will do a basic literature research on a topic/task that we provide to you. Each of those tasks covers a different topic in Brain-Computer Interaction. Those topics will be presented in the Kickoff meeting, including a short introductory lecture on different BCI techniques, providing the necessary background. After the presentation you can indicate your interest in one of those topics and we will try to assign you to a group according to your interest.
In a second step you will create a concept and implementation plan to solve the given task based on state-of-the-art methods that you found in the literature research and present your plan to the seminar participants and tutors in a 20-minute presentation. The talk should introduce the task/problem you were working on, include the most relevant studies in this field, what impact those studies and their results have on your work and finally your implementation plan that you concluded on. The implementation plan will be discussed in the group and if accepted you can proceed to the implementation.
In the third stage you will be working with the EEG data to solve the given task. We will provide you the data, the basic software components to work with it and give a tutorial on standard processing methods. Afterwards you will continue with the implementation and at the end of the seminar each group will give a short demonstration and present their results in a talk. Finally, you will write a summary of your work in the format of a typical research paper.
Due to the ongoing COV19 pandemic, the seminar will be held virtually with a possible transition to physical or hybrid meeting depending on the regulations during the semester.
Requirements: No formal requirements, however, basic knowledge in Python programming, Machine Learning and Signal processing might be helpful.
Places: 12
Build an 8-bit Computer from Scratch by Sebastian Hack
This is a practical seminar on basic computer architecture. We will design an 8-bit CPU and some peripherals from scratch and build it using 74xx series TTL logic chips.
You will:
- design your own instruction set, micro architecture and peripherals.
- Use EDA tools to create schematics (a building plan) of your machine.
- Build the machine using discrete logic chips.
- Devise techniques to simulate and test the components of your system.
- Write basic/simple system software to bring life to your machine an run simple programs.
More information: https://compilers.cs.uni-saarland.de/teaching/seminar/8bit/2022/
Requirements: Programming 1/2 and System Architecture
Places: 12
Category Theory by Derek Dreyer, Jan Menz, and Michael Sammler
Category theory is a relatively young branch of mathematics which provides a kind of "abstract theory of functions". It elucidates fundamental algebraic structures which have reappeared time and again throughout mathematics, computer science, and other disciplines, and it has proven particularly useful in organizing the foundations of logic, type theory, and programming languages. As such, anyone working on the more formal aspects of programming languages can benefit from an acquaintance with the basics of category theory.
In this seminar, we will work through an introductory textbook on category theory ("Category Theory" by Steve Awodey), covering such essential concepts as categories, functors, natural transformations, limits and colimits, functor categories, Yoneda's lemma, adjoints, and monads. Time permitting, we will also explore how category theory can be used in building the foundation of a modern, higher-order separation logic (Iris).
Requirements: An intermediate level of mathematical maturity is expected. Though there are no formal prerequisites, the course will be easiest to follow for students who have already taken Discrete Mathematics and Intro to Computational Logic (and/or Semantics).
Places: 12
Computer Vision and Machine Learning for Computer Graphics by Christian Theobalt, Marc Habermann, Thomas Leimkühler
Computer Vision strives to develop algorithms for understanding, interpreting and reconstructing information about real-world scenes from image and video data.
Computer Graphics focuses on image synthesis: algorithms to build and edit static and dynamic virtual worlds and to display them in photorealistic or stylized ways.
Machine Learning is concerned with studying and developing algorithms which use statistical models to solve problems by analyzing and drawing inference from data.
In recent years, these fields have converged more and more. Both Computer Vision and Computer Graphics create and exploit models describing the visual appearance of objects and scenes, while the most successful models heavily utilize ideas from Machine Learning. In this seminar series, we will cover advanced research topics that cross the boundaries between the fields of Computer Vision, Computer Graphics, and Machine Learning. This seminar will cover research papers from the following topics:
Image and video generation, manipulation, and analysis; multi-view geometry and reconstruction; computational photography and videography; shape matching; pose estimation, tracking, and character animation; deep learning for computer vision and computer graphics.
Seminar website: https://vcai.mpi-inf.mpg.de/teaching/vcai_seminar_2022/index.html
Requirements: This seminar is aimed at graduate students in computer science or related fields.
Places: 12
Coping with computational hardness: approximation, moderately exponential-time, and parameterized algorithms by Daniel Marx
Most optimization problems and combinatorial search problems are NP-hard, hence we do not expect to be able to find polynomial-time algorithms to solve them exactly. But even for such problems, it is still possible to prove rigorous theoretical results that show the existence of algorithms that solve the problem more efficiently than naïve or brute force approaches. Approximation algorithms do not provide an optimal solution, but there is a provable bound on the quality of solution they find. The field of moderately exponential-time algorithms designs algorithms that searches a much smaller (but still exponential) solution space than brute force algorithms do. The running time of a parameterized algorithm is polynomial in the input size, but possibly exponential (or worse!) in some well-defined parameter of the input.
Designing algorithms of this form often requires deep insights into the nature of the problem and clever algorithmic/combinatorial ideas. In this seminar, we will read, present, and discuss research results with the goal of seeing how these algorithmic paradigms can be applied to problems in different domains.
Requirements: A solid background in algorithms and some familiarity with concepts in optimization is necessary for this seminar, as well as a general affinity towards mathematical proofs.
Places: 10
Crowdsourcing high quality annotations and experimental data by Merel Scholman
Crowdsourcing observations from non-experts is one of the most common approaches to collecting data and annotations in NLP, and it is becoming increasingly popular in psycholinguistics. Crowdsourcing has been applied to a plethora of tasks, such as eliciting annotations of diverse phenomena ranging from discourse relations to image labelling, as well as obtaining experimental data such as reading times or word recognition.
Despite crowdsourcing having grown into a fundamental method for collecting data, its usage is largely guided by common practices and personal experience of researchers. This seminar has a focus on how methodology can shape our research results. We will discuss various principles and practices that have proven effective in generating high quality data for a large range of tasks.
Note: this seminar will be taught in English and will be held online via MS Teams.
We will offer this seminar twice this semester, to offer more students to take it. A second slot will be agreed on in the first session.
Places: 26
Data-driven Understanding of the Disinformation Epidemic (DUDE) by Yang Zhang
Arguably, one of the greatest inventions of humanity is the Web. Despite the fact it revolutionized our lives, the Web has also introduced or amplified a set of several social issues like the spread of disinformation and hateful content to a large number of people.
In this seminar, we will look into research that focuses on extracting insights from large corpus of data with the goal to understand emerging socio-technical issues on the Web such as the dissemination of disinformation and hateful content. We will read, present, and discuss papers that follow a data-driven approach to analyze large-scale datasets across several axes to study the multi-faceted aspects of emerging issues like disinformation.
During this seminar, the participants will have the opportunity to learn about state-of-the-art techniques and tools that are used for large-scale processing, including, but not limited to, statistical techniques, machine learning, image analysis, and natural language processing techniques.
Requirements: There are no formal prerequisites for this seminar. Despite this fact, it will be helpful if the participants have a basic understanding of statistics.
Places: 12
Datenschutz und KI: rechtliche und technische Aspekte by Prof. Dr. Sorge
Das Seminar eJustice und Datenschutz ist ein interdisziplinäres Seminar für Informatiker und Juristen.
Ablauf des Seminars:
Studierende erhalten während der Vorbesprechung ein Thema und müssen eine Seminararbeit hierzu anfertigen. Vor der Abgabe der fertigen Seminararbeit ist die Einreichung eines Abstracts vorgeschrieben. Nach der Abgabe des Abstracts kann bei Bedarf bei den Betreuern Feedback zu diesem eingeholt werden. Danach ist die Abgabe eines „Preprints“ der Seminararbeit erforderlich. Dies ist ein Vorabgabe der finalen Seminararbeit. Daraufhin findet eine Review der abgegebenen Preprints durch die weiteren Teilnehmer des Seminars statt. Jedem Studierenden werden hierfür 3-4 Paper anderer Studierender zugeteilt, welche von Ihnen reviewed werden. Die Review jedes Papers muss sich inhaltlich auf das gesamte Paper erstrecken und mindestens 400 Zeichen (für jedes Paper) enthalten. Studierende erhalten somit schon vor der endgültigen Abgabe der Seminararbeit Feedback von den anderen Teilnehmern des Seminars. Nach den Reviews finden die Vorträge statt. Abschließend werden die fertigen Seminararbeiten abgegeben.
Die Gesamtnote ergibt sich aus dem Preprint, den verfassten Reviews zu den Papern der anderen Studierenden, dem Vortrag und der finalen Seminararbeit.
Die Nichtabgabe des Abstracts, des Preprints, der Reviews oder der finalen Seminararbeit sowie ein Nichterscheinen zum Vortragstermin führt zum Nichtbestehen des Seminars.
Termine:
Vorbesprechung und Themenvergabe: 20.04.2022, 18.00 Uhr (s.t.)
Anmeldung zur Besprechung der Abstracts: 04.05.2022
Abgabe Abstracts: 11.05.2022
Abgabe Preprint: 22.06.2022
Abgabe Reviews: 06.07.2022
Vorträge: 11.07 und 15.07.2022
Finale Abgabe der Seminararbeiten: 31.07.2022
Requirements: Es wird erwartet, dass Teilnehmer in der Lage sind, Vorträgen in deutscher Sprache zu folgen.
Places: 7
Deep Learning for Drug Discovery by Michael Backenköhler, Gerrit Großmann, Verena Wolf
The discovery of novel therapeutic drugs is extremely costly and time-consuming.
In silico (i.e. computational) drug discovery can potentially accelerate this process. However, this method requires systematic exploration of the vast chemical space of possible molecules.
High expectations are placed on deep learning methods to simplify this process and provide navigation in the realm of potential drugs.
This seminar focuses on deep learning methods for learning molecular representations, predicting molecular properties, and ultimately for finding novel molecules with desirable properties.
We will focus on recent advances in the field of geometric deep learning and graph neural networks and their application for drug discovery.
https://mosi.uni-saarland.de/lectures/22_1_deep_learning_for_drug_discovery/
Requirements: Prior knowledge of biochemistry is not expected. General background knowledge in deep learning is recommended.
Places: 10
Deep Learning for Visual Computing by Pascal Peter (UdS), Patrick Trampert (ZF AI Lab)
Neural networks have significantly changed the area of image processing and computer vision, affecting both the forefront of current research and practical deployment in industry applications. Compared to early days of NN-based visual computing, not only the scope of different applications has broadened and performance has improved, but research focus has also shifted towards increased model transparency.
In this seminar, we discuss current scientific publications from the area of deep learning and investigate the corresponding models with a hands-on approach that involves implementations. We focus in particular on image inpainting, where missing or damaged image parts have to be restored from known data. This seminar is offered in cooperation with the ZF AI Tech Centre, a on campus presence of ZF in the Scheer Tower D5.2. This allows us to also take a look at industry applications such as autonomous driving and AI in manufacturing.
The seminar will be conducted online. All details can be found on the seminar website:
https://www.mia.uni-saarland.de/Teaching/dlvc22.shtml
Requirements: This seminar is for advanced bachelor or master students in Visual Computing, Computer Science, or Mathematics who are already acquainted with basic techniques of machine learning. In particular, they must posses some prior experience of implementing neural networks in python frameworks (Tensorflow/Pytorch). Such prior knowledge is required for the practical parts of the write-up. In addition, a basic level of experience in image processing and mathematics is strongly recommended.
Language: The publications discussed in this seminar are written in English, and English is the language of presentation.
Places: 12
Developing Cloud Software using SAP by Maximilian Altmeyer, Frederic Kerber, Felix Kosmalla, Christian Lander
In this seminar, we deal with Software as a Service by developing Cloud ready software. We will learn about the advantages of Cloud Computing and provide different topics for hands-on development. As one example to run such software, we will use the SAP Business Technology Platform (BTP) for deployment and use the SAP Cloud Application Programming Model (CAP, https://cap.cloud.sap/docs/).
More information can be found here: https://umtl.cs.uni-saarland.de/teaching/summer-2022/seminar-developing-cloud-software-using-sap-technology.html
Requirements: You should have passed Programming 2 or have sufficient knowledge about programming from elsewhere. Experience in user-centered design, human-computer-interaction and / or artificial intelligence is a benefit.
All meetings must be attended (virtual, exceptions require an official document, e.g. a doctor’s certificate), otherwise, you won't pass the seminar.
Places: 12
Digital and Privacy-Preserving Signatures by Lucjan Hanzlik
Digital signatures are a basic cryptographic building block that ensures authenticity (who signed) and integrity (what is signed) of messages.
In this seminar, we will learn about different types of digital signatures including not only standard RSA and ECDSA signature schemes but also recent schemes secure against quantum adversaries.
We will also take a look at signature schemes that relax the authenticity and integrity properties to increase the privacy of users and/or message like e.g.:
- ring signatures used in the Monero cryptocurrency,
- pseudonymous signatures used in the German eID, and
- blind signatures used in e-Cash.
The kick-off meeting will be during the first week of lectures. The seminar will be held in English.
Requirements: A basic understanding of cryptographic primitives such as encryption, signatures, and hash functions is required.
Places: 16
Ethical AI - AI for ethics by Dr.-Ing. Christian Müller
The current AI expansion is accompanied by constant calls for applied ethics, meant to harness the “disruptive” potentials of new AI technologies. Academia, industry, and different societal sectors are increasingly providing philosophical statements, theories, ethical guidelines, and principles aimed to advise developers through the path of producing ethical AI systems. To critically assess the implications of AI and manage its diverse impacts, the ability to articulate possible courses of action when ethically sensitive issues arise is essential. Therefore, a main objective of this seminar is to develop this capacity in an applied manner by tackling debatable questions, as: How could ethics in AI be operationalized? How are the AI ethics principles established, and upheld? How do we protect AI systems and its users against unintended consequences? Can we hold a machine liable? Who takes responsibility for an AI system error? - The developer, the company that it works for, the person/database owner from whom the system learned from? What happens if AI systems are hacked? Should we rather choose between developing bigger, faster, and fancy models and systems or instead, more transparent, fair, and free of bias systems?
Further, an explorative analysis on the way AI can improve ethics will be conducted during the discussions. To what extent, and in what form can AI itself become an effective contributor to applied ethics? Can AI enhance ethics? Could it be designed not only to accomplish the “ethical AI” requirement, but also as a tool aimed to upgrade the methods and ethical standards of different research sectors? How could the promises of such an approach be exploited, and the perils managed?
These directions will be explored, analysed, and debated during the seminar to raise awareness of AI ethical aspects, and to stimulate reflection and discussion upon implications of the use of AI in all societal sectors.
During the seminar students will:
· understand relevant concepts utilised into the AI ethics discourse, as: ethical principle, transparency, explainability, fairness, robustness, privacy, trustworthy, data governance, human-data interaction, etc.
· be sensitive to ethical issues surrounding transformative technologies and apply professional critical judgement and reflexivity to these.
· become familiar with the ethical questions and concepts related to contemporary AI capabilities.
· discover what role do AI guidelines play in shaping the discussion, and how might things develop in the future?
· explore and analyse the AI possibilities to support and improve applied ethics.
· acquire different modalities of binding ethical concepts and theories with AI by doing practical exercises
Places: 15
Ethics in Security Research by Katharina Krombholz
In this seminar, we discuss ethical considerations on cybersecurity research. We will mainly focus on empirical and offensive security research. You will learn how to identify "good" and "bad" ethical practices and how to modify research setups in order to comply with high methodological standards and ethics guidelines.
Places: 8
How to build a Social Computer by Patrick Gebhard, Cornelius König, Dimitra Tsovaltzi
Computers, mobile phones, and wearables are part of our daily lives. Its user interfaces are designed to interact with facts, e.g. health parameters, documents, etc. – which is sufficient for some applications. But what if a computer system would understand users like human do and adapt to the individual social situation in order to establish a social human-like interaction? Such concepts are highly relevant to social training systems, e.g. virtual job interview training and work-life-balance systems. This interdisciplinary seminar in Psychology and Artificial Intelligence investigates the question „How to make computers social?“. What is social? What concepts, theories help to define being social? How do humans socially communicate? How can this be transferred into a computer model? What is technically feasible? How can this be exploited for social training systems? Relevant concepts and theories will be presented, discussed, and transferred into computer models. Three prototypical interactive social computer applications will be created as proof of concept. Students of Psychology and Computer Science will work together designing, discussing, implementing, and creating an evaluation plan for each application.
More information: https://affective.dfki.de/teaching-2/social-computing-seminar/
Requirements: In total, maximal 18 students can participate in the seminar. 9 psychology students and 9 computer science students. Psychology students have the usual process for getting into seminars by registering in LSF. Computer Science students use the Seminar Assignment website by SIC.
Psychology students should have knowledge in the areas of models of emotions, models of social interaction and requirements. A master degree in Psychology seems appropriate.
Computer Science students should have knowledge in the areas of AI, HCI, and software design. A master degree in Computer Science (or equal) seems appropriate.
The main language is German, English is possible. Why? The seminar is about emotions and values that have to be discussed between psychology and computer science students. Our recommendation is that the seminar language is in German. If students agree that English is used, they should be aware that this will demand for additional efforts (e.g., in-deep explanations).
During the seminar, progress is reported at a specific web page for each project so every seminar participant can get an overview on the activities.
Project slides and short progress reports are in English!
Places: 9
Human-Centered Interface Design for Automated Cyber-Physical Systems by Anke Hirsch, Nina Knieriemen, Frederik Wiehr, Lukas Flohr
Cyber-physical systems (CPS) connect physical and virtual components in real-time.
Increasing automation transforms the role of humans in CPS from active operators to relatively passive observants who only perform an oversight task. From a human factors perspective, this results in several challenges, e.g., inappropriate trust of humans in the automated systems, insufficient situational awareness, and loss of skills (Manzey, 2012). Easy-to-understand interactions of humans and automated cyber-physical systems are crucial to counteract those issues. Therefore, it is essential to understand the interrelation between people, systems, and the environment. Human-centred design (e.g., described in ISO 9241-210) provides a straightforward framework for design.
Our topics will be from two areas: autonomous mobility-on-demand and mixed-initiative control, between a CPS and its users.
In the second area, we will apply scenarios like autonomous drones and highly automated driving (HAD). For example, the UX prototypes are concerned about generating additional multimodal explanations, with the goal to improve safety for the handover procedure.
In this seminar, you will work in small project groups supervised by one lecturer and apply collaborative UX design methods to create new UI concepts for automated CPS in a human-centered way. Based on a thoroughly conducted analysis, we will iteratively design, prototype, evaluate, and refine the concepts throughout the semester to counteract potential user acceptance hurdles and human factor challenges from early development phases.
The Seminar is planned for a hybrid mode. Regular dates (Wednesdays 14:00 – 16:00) will be conducted remotely and accompanied by two 'in-person' workshops (if the corona situation allows; see timeline).
IMPORTANT: Before signing up to the seminar, please make sure that you will be available on all dates mentioned in the timeline since attendance to both the remote and in-person seminar dates is compulsory.
Kick-Off
20 April | 14:00 – 16:00
Progress Presentation (Research & Analysis | Online)
11 May 14:00 – 16:00
18 May 14:00 – 16:00
Ideation Workshop @Ergosign
25 May 09:00 – 16:00
Progress Presentation (Prototyping & Evaluation Plan)
13 July 14:00 – 16:00
20 July 14:00 – 16:00
Final Presentation @Ergosign
28 September 09:00 – 16:00
Final Report Submission Deadline
28 September 23:59
Details can be found on the seminar website:
https://umtl.cs.uni-saarland.de/teaching/summer-2022/seminar-human-centered-interface-design-for-automated-cyber-physical-systems.html
Requirements: To participate you should have
* basic knowledge in Human-Computer Interaction (passed core lecture)
* some experience with UI prototyping tools (e.g., Sketch, Figma, Botmock) OR good programming skills in web development, if you prefer to develop your prototype.
* This seminar aims primarily at master students in Computer Science who preferably hold a B.Sc. degree in this or related field.
* be present in Saarbrücken
Places: 12
Machine Learning for Social Signal Processing by Philipp Müller, Hali Lindsay, Chirag Bhuvaneshwara, Fabrizio Nunnari
Social signals like eye contact, prosody, facial expressions, and body language are central to human social life. In this seminar, we will explore how machine learning can be used to detect and interpret such signals.
Participants will form small groups and can choose from a number of possible projects, including:
• Eye contact detection
• Inferring leadership, rapport, or liking from social signals in group interactions
• Detecting backchanneling behaviour
• Detecting medical conditions from speech behaviour
To keep projects inside a reasonable scope for a seminar and to allow students to focus on the development of machine learning approaches, we will provide well-documented datasets, annotations, as well as relevant pre-computed input features.
Seminar Website: https://affective.dfki.de/teaching-2/machine-learning-for-social-signal-processing-ss2022/
The first (virtual) session will be on 12 of April at 14:00 via MS Teams.
Requirements: • The seminar is targeted at master students interested in pursuing research in the social signal processing field
• Knowledge in Machine Learning (e.g. Machine Learning core lecture) and preferably also speech processing and computer vision
• Practical experience with scientific computing in Python
Places: 12
Neural-Symbolic Computing by Bernd Finkbeiner and Frederik Schmitt
The way our brain forms thoughts can be classified into two categories (according to Kahneman in his book “Thinking Fast and Slow”):
System 1: fast, automatic, frequent, stereotypic, unconscious. Is this a cat or a dog? What does this sentence mean in English?
System 2: slow, effortful, logical, conscious. 17*16 = ? If a -> b does b -> a?
The traditional view is that deep learning is limited to System 1 type of reasoning. Mostly because of the perception that deep neural networks are unable to solve complex logical reasoning tasks reliably. Historically, applications of machine learning were thus often restricted to sub-problems within larger logical frameworks, such as resolving heuristics in solvers. In this seminar, we will explore new research that shows that deep neural networks are, in fact, able to reason on “symbolic systems”, i.e., systems that are built with symbols like programming languages or formal logics.
Example Topics: What are your chances against an AI in a programming competition? Is it possible to teach temporal logics to neural networks? Can neural networks discover unknown connections in mathematics?
Requirements: Participants should have strong interest in logical reasoning and/or machine learning. There is, however, no formal prerequisite.
Places: 12
Neuro-Symbolic Reinforcement Learning by Adish Singla
The course will provide an overview of the research in Neuro-Symbolic Reinforcement Learning. The course consists of three main components as follows:
(1) Research papers: During the first half of the semester, students will be assigned about 6-8 research papers and will have to write a short report for each paper. These reports will be due during the semester, about one report per week.
(2) Project: During the second half of the semester, students will work on an implementation project.
(3) Final presentation: At the end of the semester, students will give a 25 mins presentation for one of the papers or the project.
This course is offered as a block seminar, and there will be no weekly classes.
Requirements: There are no formal requirements; however, please note the following points:
(1) Preference will be given to students who have already taken courses covering topics such as machine learning, artificial intelligence, reinforcement learning, program analysis, software testing, and system security.
(2) We are requesting you to provide a short motivation letter (about 10 lines) explaining the reasons why you are interested in taking this seminar. Importantly, you should mention all the relevant courses that you have taken along with your grades. Please provide this information in the text box below.
Places: 10
Psychological Theories for Intelligent and Interactive Systems by Anna Maria Feit, Markus Langer
When do we trust an AI system; what's our search strategy when we google something; how does the presentation of different options influence our decision (e.g. when booking a hotel or taking an online bet)?
In this Seminar, we will answer such question by learning about fundamental theories of psychology and how they can be applied to the design of intelligent and interactive systems.
Each session will feature another theory that aims to explain human behavior for example around decision making, trust, risk taking, motivation and many other. We will discuss concrete examples of how these theories can be applied in human-computer interaction and develop our own ideas for utilizing them in the design of intelligent systems and experimental studies.
The organizational details will be published soon.
This interdsiciplinary seminar is also open to students from Psychology, which will surely make for interesting discussions and cross-fertilization.
Requirements: No specific requirements. Students should have an interest in learning about the human component of interactive systems.
Places: 8
Quantum Machine Learning by Matthias Klusch, Antonio Macaluso
Quantum Machine Learning (QML) is an emerging field that aims to leverage the properties of quantum computation to improve traditional machine learning algorithms. In this seminar, we will take a closer look at selected methods of quantum machine learning and discuss their strengths and weaknesses.
The seminar type is classic in the sense that registered participants will present assigned topics and discuss the strength and weaknesses of presented approaches. In addition, there will be two dedicated opponents for each presentation of an assigned topic.
Seminar website: http://www.dfki.de/~klusch/QML-ss22
Due to the ongoing COV19 pandemics, the seminar will be held virtually;
later transition to physical or hybrid meeting possible.
Requirements: This seminar aims primarily at master students in Computer Science who preferably hold a B.Sc. degree in this or related field. Good knowledge in AI (introductory course on AI), machine learning including deep learning, and mathematics (in particular linear algebra) is required.
Places: 8
Reading Group: String Algorithms by Karl Bringmann, Tomasz Kociumaka, Alejandro Cassis, Nick Fischer
This seminar will discuss recent algorithms for text processing and sequence similarity, as published in conferences like STOC, FOCS or SODA. We study sequence similarity measures such as edit distance and longest common subsequence, which are ubiquitous in text processing and bioinformatics. Starting from the basic dynamic programming algorithms that run in quadratic time, we study approximation algorithms that run in near-linear time or even sublinear time. We also cover approximate pattern matching as well as embedding, sketching, and possibly streaming algorithms.
The organizers bring expertise on different aspects of these algorithms. One goal of this course is that the organizers teach each other what they know about string algorithms. The seminar will therefore feature about 10 talks by the organizers, presenting their work on string algorithms, in a format similar to the Algorithms and Data Structures core lecture. Additional to attending these lectures, your task is to read and present one paper on string algorithms, and possibly write a summary.
For more details see the course website (to be created soon): https://cms.sic.saarland/strings22/
Requirements: The core lecture Algorithms and Data Structures is a mandatory requirement. In exceptional cases we can allow students who do not fulfill this requirement but have attended other relevant courses (e.g. basic lecture on Algorithms and Data Structures, Algorithms for Sequence Analysis).
In the motivation text below, let us know whether you have passed the core lecture Algorithms and Data Structures, and list other relevant coursework (e.g. basic lecture on Algorithms and Data Structures, Algorithms for Sequence Analysis, special lectures on algorithm theory). Also please describe in 1-2 sentences your motivation for participating in this seminar.
Places: 5
Software Engineering Research in the Neuroage by Sven Apel, Annabelle Bergum, Thomas Bock, Sebastian Boehm, Christian Hechtl, Christian Kaltenecker, Norman Peitek, Florian Sattler, Kallistos Weis
The pivotal role of software in our modern world imposes strong requirements on quality, correctness, and reliability of software systems. The ability to understand program code plays a key role for programmers to fulfill these requirements. Despite significant progress, research on program comprehension has had a fundamental limitation: program comprehension is a cognitive process that cannot be directly observed, which leaves considerable room for (mis)interpretation, uncertainty, and confounding factors. Thus, central questions such as “What makes a good programmer?” and “How should we program?” are surprisingly difficult to answer based on the state of the art.
Recently, researchers began to lift research on program comprehension to a new level. The key idea is to leverage recent methods from cognitive neuroscience to obtain insights into the cognitive processes involved in program comprehension. Opening the “black box” of human cognition will lead to a breakthrough in understanding the why and how of program comprehension and to a completely new perspective and methodology of measuring program comprehension, with direct implications for programming methodology, language design, and education.
In this seminar, we will review and discuss the past, current, and future developments in this area.
In this seminar, each participant has to perform a literature search and propose an experiment for the given topic.
Subsequently, the topic, the results of the literature search, and the proposed experiment have to be incorporated into a presentation and a written thesis.
To aid the literature search, the experiment proposal, and the presentation, this seminar includes multiple preparatory sessions at the beginning of the semester.
The student presentations will be held in June and July 2022.
All sessions will take place on-site at the university (under the caveat that the pandemic situation admits in-person sessions) on Thursdays 12:15 PM - 2:00 PM.
Participation to all sessions is mandatory.
The topic assignment will take place on Thursday April 21, at 12:15 PM. Further information will be provided via e-mail after registration.
Requirements: Basic knowledge on software engineering and programming.
Places: 9
Symbolic Execution (Multimodal Seminar) by Dominic Steinhöfel
Symbolic execution is a powerful program analysis technique based on the idea to execute a program with symbolic placeholders instead of concrete inputs. Since this principle enables exploring many program paths at once, it has been used for intensive software testing and full program proving. The most significant obstacles symbolic execution is facing are path explosion (a new program path is spawned at every case distinction) and the overhead due to constraint solving. In the focus of the seminar are the basic principles of symbolic execution and approaches for mitigating the aforementioned problems.
Only five students can participate in this intensive seminar.
Organization
This seminar consists of ten regular sessions held during the semester and a final block session. Each regular session is either a lecture, paper, or lab session.
In lecture sessions, the students are given an introduction into general topics (giving talks) and the specific course topic (symbolic execution).
In paper sessions, one or two students present different papers on the topic of symbolic execution. These presentations only take five minutes and are not graded. The purpose of this is to reflect on presentation style and to provide a high-level overview for kickstarting the upcoming lab session(s). In advance of each paper session, each student (regardless of whether or not they present a paper in that session) is requested to submit a brief summary of the paper.
Lab sessions aim to solidify the gained theoretical knowledge by practical application. Each student is supposed to solve an appropriately sized programming exercise related to the topic of the lab session (and the preceding paper session) in advance of the session. In the session itself, one student presents their solution and the gained insights in a short presentation; afterward, a discussion of the implemented technique follows. After the regular sessions, each student will have presented one paper and one lab session in brief talks, and will have submitted summaries for up to five (but no less than four) papers.
Finally, each student presents theoretical and practical aspects of one symbolic execution topic in a final block session. The topics are those of the papers discussed in the regular sessions; but each student shall present a different topic than the one they have already presented.
Grading
The grade is built from the final talk (50%) and all submitted lab solutions (50%, i.e., 10% for each submitted solution). Each student has one “joker” for the lab solutions: The worst individual grade for a lab solution (including missed submissions) is replaced by the best obtained such grade. To pass the seminar, it is mandatory to (1) attend all seminar sessions, unless a student is absent due to illness, (2) submit at least four short summaries, (3) give one short talk on a paper and one on a lab topic, (3) submit at least four lab solutions, (4) give a final presentation, and (5) obtain at least 50% of the maximum number of points that can be achieved in the final talk and the lab solutions.
For details, see the seminar home page at https://cms.cispa.saarland/symex_22/
Requirements: Solid knowledge of programming languages and their semantics is a must, as indicated by a BSc degree in computer science or a related field, passing "Programming 2", or similar knowledge.
While not a formal prerequisite, a relatively solid knowledge of Python is highly recommended for the lab sessions, or otherwise should be acquired until one week before the first lab session.
Places: 5
Virtual Reality Game Development using a Rock Climbing Treadmill by Maximilian Altmeyer, Florian Daiber, Donald Degraen, Felix Kosmalla
In this seminar, students will have the opportunity to work on a physical climbing treadmill that was integrated into a virtual environment, allowing for immersive climbing. The students will use this platform as a virtual reality rock climbing exergame environment.
The aim of the seminar is to develop exergames using this platform in small groups of students. This entails a conceptualization phase of a game which should be aligned to one of the topics outlined below. After initial feedback, the games should be developed. For this, we provide physical access to the climbing wall itself and VR hardware that can be used on DFKI premises. A simulator will be made available for remote work.
More information can be found here: https://umtl.cs.uni-saarland.de/teaching/summer-2022/seminar-virtual-reality-game-development-for-a-rock-climbing-treadmill.html
Requirements: - Strong background in Unity 3D development
- Willingness to think out of the box and produce own ideas
- Your own laptop or workstation that is capable of running Unity 3D
- Optional, but helpful:
* experience in VR development
* 3D modeling skills
* has visited a climbing gym at least once
Places: 12