- informatik-vortraege - lists.rwth-aachen.de

Einladung: Informatik-Oberseminar Sabarinath Mahadevan
by Mahadevan, Sabarinath 21 May '24

21 May '24

+********************************************************************** * * * Einladung * * * * Informatik-Oberseminar * * * +********************************************************************** Zeit: Freitag, 24. Mai 2024, 10.30 Uhr Ort: Raum 025, Mies-van-der-Rohe Str. 15 (UMIC Gebäude) Der Vortrag findet hybrid statt. Zoom: https://rwth.zoom-x.de/j/64536789621?pwd=d2I5Y1VDM2xscVVDbTF0ZnJVWUUvZz09 Meeting ID: 645 3678 9621, Passcode: 998171 Referent: Sabarinath Mahadevan, M.Sc. Lehrstuhl Informatik 13 Thema: The Many Facets of Object Segmentation in Images and Videos Abstract: Segmentation is an important task in computer vision where the pixels belonging to a region of interest, that often share similar characteristics, have to be separated out, and assigned a unique label. This task is relevant in both the image and video domains, and finds direct applications in a wide range of fields including but not limited to autonomous driving, robotics, image editing, and surveillance. For both images and videos, the task of segmentation has various flavours that are often highly related and equally challenging. With the advent of deep learning, modern computer vision algorithms are able to leverage large amounts of available data to achieve impressive performance for many of these segmentation problems. However, such algorithms are often catered either towards segmenting a pre-defined set of object categories, or to some specific sub-domains, and require to be adapted for out-of-domain tasks by training these methods on additional domain-specific data, that is expensive to annotate. As a result, we need algorithms that generalise well to data from unseen domains and also to new task settings in addition to having methods that can efficiently annotate data. The first part of this thesis focusses on efficiently annotating objects using Interactive Segmentation, where the goal is to segment and refine objects in an image using user clicks. Here, I'll present two of our research works that advances the state-of-the-art in the interactive segmentation domain. The first among these, ITIS, develops a novel iterative training strategy in which clicks are added iteratively during training based on the error regions in the network predictions. The iterative training strategy aligns the simulated user-click patterns that are used during training with the actual click patterns that the network would encounter at test time. While this strategy is effective in reducing the number of clicks required to annotate objects, ITIS can only segment one object at a time due to the limitations of its network architecture. We address this problem in our subsequent work called DynaMITe, where we formulate user clicks as spatiotemporal sequences, and develop a novel Transformer based formulation that can process such a sequence and encode them into relevant object or region descriptors. These descriptors are then used to generate the relevant instance segmentation masks. Unlike previous methods, our architecture can process clicks for multiple objects at once, and correspondingly predicts non overlapping segmentation masks without any post-processing. In the second part of this talk, I'll focus on end-to-end video segmentation networks based on 3D convolutions, and present our work STEm-Seg. STEm-Seg is a a bottom-up end-to-end approach for instance segmentation on videos, which uses a partially 3D network to learn spatio-temporal embeddings that can be clustered into instance tubes based on the predicted clustering parameters. Our method is very generic and can be applied to a wide range of instance segmentation tasks in videos. Finally, I’ll present our latest work, Point-VOS, where we show that video segmentation models can learn from spatiotemporally sparse point annotations instead of dense per-object mask annotations. We also present an efficient point-wise annotations scheme, and use it to annotate two large-scale video datasets with associated language expressions. We also present a new Point-VOS benchmark and the corresponding baselines, and show that on our point annotations can be used to achieve results close to state-of-the-art models that use dense mask supervision. Additionally, we evaluate models that connect vision and language on the VNG task, and correspondingly show that our data helps in improving their performance.

1 0

UnRAVeL Survey Lecture - Today!
by Kassing, Jan-Christoph 16 May '24

16 May '24

Dear all, this is a reminder for Martin Grohe's talk with the title "The Complexity of Constraint Satisfaction" taking place today (16.05) at 12:30 in the B-IT room 5053.2. Please find the details below --- Abstract --- The objective of a constraint satisfaction problem (CSP) is to assign values to variables subject to constraints on the values. Obviously, this is a very general type of problem, and it is not surprising that many algorithmic problems in various application domains can be described as CSPs. It is neither surprising that, in general, CSPs are computationally hard. Considerable efforts have been made to precisely understand the complexity of CSPs, with the goal of identifying tractable restrictions and, ultimately, determining the boundary between tractable and intractable CSPs. A deep theory based on methods from logic, combinatorics, and universal algebra has evolved around these questions. In this survey talk, I will give an overview of the main results and techniques of this theory, and I will highlight some current research topics and open questions. ---------------- Part of the programme of the research training group UnRAVeL is a series of lectures on the topics of UnRAVeL’s research thrusts algorithms and complexity, verification, logic and languages, and their application scenarios. Each lecture is given by one of the researchers involved in UnRAVeL. This years topic is "UnRAVeL - New Ideas!". In these lectures, UnRAVeL professors will discuss current research as well as highlight open problems and offer a perspective on potential future directions. All interested doctoral researchers and master students are invited to attend the UnRAVeL lecture series 2024 and engage in discussions with researchers and doctoral students. We are looking forward to seeing you at the lectures. Kind regards, Jan-Christoph for the organisation committee

1 0

Einladung: Informatik-Oberseminar Barna Zajzon
by Abigail Morrison 06 May '24

06 May '24

+********************************************************************** * * * Einladung * * * * Informatik-Oberseminar * * * +********************************************************************** Zeit: 08.05.2024 um 09:00 Ort: Informatikzentrum, Raum 9222 in E3 Referent: Barna Zajzon, M.Sc. Lehrstuhl Informatik 3 Thema: Sequential information processing in modular spiking networks Abstract: Established paradigms formulate many cognitive processes in terms that involve the manipulation of sequentially organized time-discrete (symbolic) representations. This underscores two basic functional requirements that cortical circuits must fulfill: the ability to create suitable representations from a highly volatile and noisy environment; and the capacity to process, and learn from, their spatio-temporal structure. Combining software tools, simulation studies and theoretical analysis, this thesis entails a series of research projects with the shared goal of disentangling how modular structures enable neural circuits to learn and process sequential information in an efficient and reliable manner. The first part analyses the characteristics of state representations in modular spiking networks and the architectural and dynamical constraints that influence the system's ability to retain, transfer and integrate stimulus information in the presence of noise. It explores the novel hypothesis that modular topographic maps, a pervasive anatomical feature of the cortex, may provide a structural scaffold for sequential denoising of stimulus representations. The second part of this work is dedicated to investigating existing, biologically detailed models of sequence processing. If we are to harvest the knowledge within these models and arrive at a deeper mechanistic understanding of the involved phenomena, it is critical that the models and their findings are accessible, reproducible, and quantitatively comparable. We illustrate these aspects through a replication study and then lay the initial steps towards a conceptual and practical, theoretically-grounded framework for benchmarking and comparison of biophysical sequence learning models. Es laden ein: die Dozentinnen und Dozenten der Informatik -- Prof. Dr. Abigail Morrison Institute for Advanced SImulation (IAS-6) Jülich Research Center & Computer Science 3 - Software Engineering RWTH Aachen http://www.fz-juelich.de/inm/inm-6 http://www.se-rwth.de Office: +49 2428 8097504 Fax # : +49 2461 61-9460 Pronouns: she/her ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Forschungszentrum Jülich GmbH 52425 Jülich Sitz der Gesellschaft: Jülich Eingetragen im Handelsregister des Amtsgerichts Düren Nr. HR B 3498 Vorsitzender des Aufsichtsrats: MinDir Stefan Müller Geschäftsführung: Prof. Dr. Astrid Lambrecht (Vorsitzende), Karsten Beneke (stellv. Vorsitzender), Dr. Ir. Pieter Jansens ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------

1 0

UnRAVeL Survey Lecture - Today!
by Kassing, Jan-Christoph 02 May '24

02 May '24

Dear all, this is a reminder for Jürgen Giesl's talk with the title "Termination and Complexity Analysis of (Probabilistic) Programs: Results and Future Work" taking place today at 12:30 in the B-IT room 5053.2. Please find the details below --- Abstract --- We give an overview on our work on developing techniques for automated termination and complexity analysis for different kinds of programs. While automated tools for the analysis of non-probabilistic programs are already very powerful, we will indicate several research problems that have to be solved in order to obtain techniques of similar power for probabilistic programs. ---------------- Part of the programme of the research training group UnRAVeL is a series of lectures on the topics of UnRAVeL’s research thrusts algorithms and complexity, verification, logic and languages, and their application scenarios. Each lecture is given by one of the researchers involved in UnRAVeL. This years topic is "UnRAVeL - New Ideas!". In these lectures, UnRAVeL professors will discuss current research as well as highlight open problems and offer a perspective on potential future directions. All interested doctoral researchers and master students are invited to attend the UnRAVeL lecture series 2024 and engage in discussions with researchers and doctoral students. We are looking forward to seeing you at the lectures. Kind regards, Jan-Christoph for the organisation committee

1 0

UnRAVeL Survey Lecture - Today!
by Kassing, Jan-Christoph 25 Apr '24

25 Apr '24

Dear all, this is a reminder for Nils Nießen's talk with the title "Open (Research) Problems in Railways. What to do?" taking place today at 12:30 in the B-IT room 5053.2. Please find the details below --- Abstract --- Today, many problems exist in railway operations. Travelling by train is associated with uncertainty: Rail passengers often complain about delayed trains. In future, however, many more trains will be running on the network. What are the causes of these problems and how can they be tackled in future? What research approaches exist to ensure stable operations? In this interactive lecture, open problems from the railway sector will be identified and discussed together. ---------------- Part of the programme of the research training group UnRAVeL is a series of lectures on the topics of UnRAVeL’s research thrusts algorithms and complexity, verification, logic and languages, and their application scenarios. Each lecture is given by one of the researchers involved in UnRAVeL. This years topic is "UnRAVeL - New Ideas!". In these lectures, UnRAVeL professors will discuss current research as well as highlight open problems and offer a perspective on potential future directions. All interested doctoral researchers and master students are invited to attend the UnRAVeL lecture series 2024 and engage in discussions with researchers and doctoral students. We are looking forward to seeing you at the lectures. Kind regards, Jan-Christoph for the organisation committee

1 0

UnRAVeL Survey Lecture - Today!
by Kassing, Jan-Christoph 18 Apr '24

18 Apr '24

Dear all, this is a reminder for Erika Ábrahám's talk with the title "Relating Stochastic Hybrid Petri Nets and Stochastic Hybrid Automata" taking place today at 12:30 in the B-IT room 5053.2. Please find the details below --- Abstract --- Hybrid systems are systems with mixed discrete-continuous behavior, typical examples being continuously evolving physical systems controlled by discrete controllers. If such systems also posses some stochastic behavior, then we talk about stochastic hybrid systems. In this talk we focus on the question how to model a group of such systems, which evolve concurrently. Different modeling formalisms offer different views on concurrent stochastic hybrid systems. On the one hand, (discrete) Petri nets, which model concurrency in a deeply inherent way, have been extended with continuous and stochastic components. On the other hand, finite automata has been extended with continuous evolution to hybrid automata, and different approaches have been proposed to integrate also stochastic components to hybrid automata. In the first part of this talk, we will have a closer look on the modeling of hybrid systems within these formalisms, and discuss an open problem, which sounds simple, but it seems to be hard to solve. In a second part, we will turn our attention to adding stochasticity to these modeling formalisms. We will discuss and relate existing approaches, make an attempt to find motivations for different design choices, and conclude with some general observations. ---------------- Part of the programme of the research training group UnRAVeL is a series of lectures on the topics of UnRAVeL’s research thrusts algorithms and complexity, verification, logic and languages, and their application scenarios. Each lecture is given by one of the researchers involved in UnRAVeL. This years topic is "UnRAVeL - New Ideas!". In these lectures, UnRAVeL professors will discuss current research as well as highlight open problems and offer a perspective on potential future directions. All interested doctoral researchers and master students are invited to attend the UnRAVeL lecture series 2024 and engage in discussions with researchers and doctoral students. We are looking forward to seeing you at the lectures. Kind regards, Jan-Christoph for the organisation committee

1 0

UnRAVeL Survey Lecture: New Ideas!
by Kassing, Jan-Christoph 11 Apr '24

11 Apr '24

Dear all, this is a reminder for Joost-Pieter Katoen's talk with the title "Facing Uncertainty in AI: From Verification to Synthesis" taking place today at 12:30 in the B-IT room 5053.2. Please find the details below --- Abstract --- Uncertainties occur in different forms: data may be noisy, mechanisms may be inherently randomised, the visibility (of e.g. a robot) may not be optimal, and the environment in which a system needs to operate may behave in an unknown manner. The central question that we will address is "Can we guarantee that AI systems are safe and dependable in the presence of such uncertainty?'' We advocate using model-based, formal verification and synthesis with a particular focus on automation. We will present techniques to verify uncertainty aspects modeled as randomness and to use formal synthesis to complete partial designs. Several example AI systems will illustrate the capabilities of these approaches. ---------------- Part of the programme of the research training group UnRAVeL is a series of lectures on the topics of UnRAVeL’s research thrusts algorithms and complexity, verification, logic and languages, and their application scenarios. Each lecture is given by one of the researchers involved in UnRAVeL. This years topic is "UnRAVeL - New Ideas!". In these lectures, UnRAVeL professors will discuss current research as well as highlight open problems and offer a perspective on potential future directions. All interested doctoral researchers and master students are invited to attend the UnRAVeL lecture series 2024 and engage in discussions with researchers and doctoral students. We are looking forward to seeing you at the lectures. Kind regards, Jan-Christoph for the organisation committee

1 0

UnRAVeL Survey Lecture: New Ideas!
by Kassing, Jan-Christoph 08 Apr '24

08 Apr '24

Dear all, part of the programme of the research training group UnRAVeL is a series of lectures on the topics of UnRAVeL’s research thrusts algorithms and complexity, verification, logic and languages, and their application scenarios. Each lecture is given by one of the researchers involved in UnRAVeL. This years topic is "UnRAVeL - New Ideas!". In these lectures, UnRAVeL professors will discuss current research as well as highlight open problems and offer a perspective on potential future directions. All interested doctoral researchers and master students are invited to attend the UnRAVeL lecture series 2024 and engage in discussions with researchers and doctoral students. All events take place on Thursdays 12:30 to 14:00, Computer Science Center, Building E2, ground floor, B-IT room 5053.2. The schedule is as following: 11.04 - Joost-Pieter Katoen: Facing Uncertainty in AI: From Verification to Synthesis 18.04 - Erika Ábrahám: Relating Stochastic Hybrid Petri Nets and Stochastic Hybrid Automata 25.04 - Nils Nießen: Open (Research) Problems in Railways. What to do? 02.05 - Jürgen Giesl: Termination and Complexity Analysis of (Probabilistic) Programs: Results and Future Work 16.05 - Martin Grohe: The Complexity of Constraint Satisfaction 06.06 - Sebastian Trimpe: Bayesian Optimization for High-Dimensional, Adaptive, and Safe Controller Learning 13.06 - Christina Büsing: Robust Optimization in Health Care 27.06 - Michael Schaub: How can algebraic topology help with data analysis? 04.07 - Gerhard Lakemeyer: Challenges in Cognitive Robotics 11.07 - Christopher Morris: Understanding the Generalization Abilities of Graph Neural Networks: Current Results and Future Directions 18.07 - Britta Peis: Future Research in Submodular Function Optimization We are looking forward to seeing you at the first lecture this Thursday! Kind regards, Jan-Christoph for the organisation committee

1 0

Einladung zum Informatik-Oberseminar am 08.04.2024
by Heinrichs, Benedikt 22 Mar '24

22 Mar '24

+********************************************************************** * * * Einladung * * * * Informatik-Oberseminar * * * +**************************************************************** Zeit: Montag, 8. April 2024, 13.30 Uhr Ort: Seminarraum 3, Kopernikusstraße 6 Referent: Benedikt Heinrichs M.Sc. IT Center, RWTH Aachen University Thema: Asynchronous Tracking and Description of Research Data Changes in Distributed Systems with Interoperable Metadata Abstract: With the digital revolution, the way to approach research has fundamentally changed. Suddenly, research processes created digital research data that needed to be stored. Initially, no standards for this existed, so practices diverged wildly. Consequently, data was produced that was not findable without a management system. For this reason, movements entered the picture intending to standardize these processes and define how research data should be managed. One recommendation is the FAIR Guiding Principles, which describe that research data should be findable, accessible, interoperable, and reusable. While these principles have set goals, no implementation guideline is provided since the different research areas are too diverse. Therefore, research data management (RDM) teams around the globe have created numerous implementations. Some of them are platforms like Coscine, which can manage research data and try to adhere to parts of the FAIR principles. However, such platforms face the issue that researchers want to store their research data with an enterprise-ready and openly accessible storage provider. Therefore, research data often does not move through these platforms but directly through the storage providers. This circumstance contradicts the aim of following the FAIR principles because the platforms cannot account for the research data movement and miss critical provenance information. The presented thesis aims to close that gap by providing a method to calculate the missing provenance information after changes occur. This so-called asynchronous data provenance is produced by comparing representations of research data. If the representations have changed, a new version or variant of the research data has likely been created. Representations can range from a generated hash to interoperable metadata about the research data. This interoperable metadata is created by running a pipeline that receives research data and extracts valuable information about its content. This information is annotated as interoperable metadata by following existing application profiles and ontologies. Interoperable metadata can be used to compute the similarity of research data with a method called FSS Jaccard. The created methods are integrated into a standards-based RDM system (RDMS), defined in this thesis, to show their applicability. For this standards-based RDMS, Coscine is used as a use case. Thereby, this thesis presents a method that can provide additional information about research data and close the presented gap for any standards-based RDMS. By using this method, RDM teams can come closer to supporting the implementation of the FAIR principles and improving the processes for researchers. Es laden ein: die Dozentinnen und Dozenten der Informatik

1 0

Einladung: Informatik-Oberseminar Jonathon Luiten
by Luiten, Jonathon 04 Mar '24

04 Mar '24

+********************************************************************** * * * Einladung * * * * Informatik-Oberseminar * * * +********************************************************************** Zeit: Dienstag, 6. März 2024, 13.00 Uhr Ort: Raum 025, Mies-van-der-Rohe Str. 15 (UMIC Gebäude) Referent: Jonathon Luiten, M.Sc. Lehrstuhl Informatik 13 Thema: Dynamic 3D Representations and Robust Evaluation for Visual Tracking Abstract: Visual tracking is a core task within computer vision, one that involves understanding the motion and persistence of the dynamic world when observed in video. Building performant tracking algorithms is critical for many applications such as robotics; self-driving vehicles; virtual and augmented reality; scene-analysis for sporting, retail and construction scenarios; and content creation and editing. While other areas of computer vision such as recognition and detection have recently reached outstanding performance due to deep learning and extremely large training datasets, tracking has remained an incredibly difficult task where such approaches have not been able to achieve similar success. We argue that tracking is inherently different from these tasks and that simply scaling up compute and data is not going to be enough. In this thesis we develop what we believe to be the missing piece holding tracking back from similar success: the use of dynamic 3D representations that can be used to model the underlying scene. Furthermore, we find that the second thing holding back the field of visual tracking was the lack of adequate evaluation metrics and benchmark settings. We address these limitations by introducing novel metrics and benchmarks, which are crucial for measuring the performance of algorithms and guiding the field toward making meaningful progress. The first half of this thesis deals with approaches to lift representations for tracking to 3D, both at the level of whole objects (MOTSFusion) and at the level of infinitesimal 3D scene elements (Dynamic 3D Gaussians). Traditionally, tracking involves finding correspondences between static 2D representations in each timestep, such as pixels or bounding-boxes. Instead, we represent the world as a set of dynamic 3D representations that move around over time in order to consistently represent the same physical location in space as it moves. We reformulate tracking from a correspondence estimation problem, to an analysis-by-synthesis problem of fitting an underlying dynamic 3D model, whose motion explains changes in image content across timesteps. By using 3D representations we can better model appearance changes due to the 3D motion of the scene and the motion of the camera through the world, while also making use of intuitive physics knowledge about how objects move through the 3D world. This enables us to both obtain better tracking results, while also resulting in consistent dynamic 3D representations that are directly useful for many downstream tasks. The second half of this thesis deals with building robust metrics and benchmarks for evaluating the performance of visual tracking algorithms. For the task of Multi-Object Tracking (MOT), previous evaluation metrics have been sorely lacking, focusing only on particular aspects of tracking performance (e.g. detection or association), but not being able holistically measure improvements in tracking performance. Furthermore, tracking evaluation has been limited to settings where only a small number of fixed object classes were evaluated. We address both of these evaluation limitations by proposing the HOTA Metrics for evaluating tracking performance in a fair and holistic way, and introducing the task of Open-World Tracking for extending tracking evaluation to a open-world setting where a potentially unlimited set of object classes need to be tracked, even if they were not previously seen during training. Together, these mark a step-change in how tracking methods are evaluated and benchmarked, and allow the tracking community to make meaningful progress towards more performant and useful tracking algorithms. Overall, by developing both dynamic 3D representations for tracking and a novel set of evaluation metrics and benchmarks, this thesis provides a number of crucial missing pieces that are needed to move towards truly useful and performant tracking algorithms, and thus toward the success of the multitude of applications for which tracking is a core component. Es laden ein: die Dozentinnen und Dozenten der Informatik

1 0