Review: Why AI is shaping our games [Video][Slides]

The review of the TEWI colloquium of Dr. Johanna Pirker from May 16, 2019 comprises the video and slides (below):

Abstract:

AI is used to create parts of our games. It provides intelligent enemy behavior, techniques such as pathfinding or can be used to generate in-game content procedurally. AI can also play our games. The idea to train computers to beat humans in game-like environments such as Jeopardy!, Chess, or soccer is not a new one. But can AI also design our games? The role of Artificial Intelligence in the game development process is constantly expanding. In this talk, Dr. Pirker will talk about the importance of AI in the past, the present, and especially the future of game development.

Bio:

Dr. Johanna Pirker is researcher at the Institute of Interactive Systems and Data Science at Graz University of Technology (TUG). She finished her Master’s Thesis during a research visit at Massachusetts Institute of Technology (MIT) working on collaborative virtual world environments. In 2017, she finished her doctoral dissertation in computer science on motivational environments under the supervision of Christian Gütl (TUG) and John Belcher (MIT). She specialized in games and environments that engage users to learn, train, and work together through motivating tasks. She has long-lasting experience in game design and development, as well as virtual world development and has worked in the video game industry at Electronic Arts. Her research interests include AI, data analysis, immersive environments (VR), games research, gamification strategies, HCI, e-learning, CSE, and IR. She has authored and presented numerous publications in her field and lectured at universities such as Harvard, Berlin Humboldt Universität, or the University of Göttingen. Johanna was listed on the Forbes 30 Under 30 list of science professionals.

Posted in TEWI-Kolloquium | Leave a comment

Review: Artificial Intelligence (AI) in media applications and services [Video][Slides]

The review of the TEWI colloquium of Dr.-Ing. Christian Keimel  from May 9, 2019 comprises the video and slides (below):

Abstract: Artificial Intelligence (AI) is nowadays used frequently in many application domains. Although sometimes considered only as an afterthought in the public discussion compared to other domains such as health, transportation, and manufacturing, the media domain is also transformed by AI enabling new opportunities, from content creation e.g. “robojournalism” and individualised content to optimisation of the content production and distribution. Underlaying many of these new opportunities is the use of AI in its current reincarnation as deep learning for understanding the audio-visual content by extracting structured information from the unstructured data, the audio-visual content.

In this talk the current understanding and trends of AI will therefore be discussed, what can be done, what is done, and what challenges remain in the use of AI especially in the context of media applications and services. The talk is not so much focused on the details and fundamentals of deep learning, but rather on a practical perspective on how recent advances in this field can be utilised in use-cases in the media domain, especially with respect to audio-visual content and in the broadcasting domain.

Bio: Christian Keimel received his B.Sc and Dipl.-Ing.(Univ.) in information technology from the Technical University of Munich (TUM) in 2005 and 2007, respectively. In 2014 he received a Dr.-Ing. degree from TUM for his dissertation on the “Design of video quality metrics with multi-way data analysis.” Since 2013 he is with the Institut for Rundfunktechnik (IRT), the research and competence centre of the public service broadcasters of Austria, Germany, and Switzerland, where he leads the machine learning team, working on the applications of machine learning and AI in the broadcasting context. In addition, he is a lecturer at TUM for “Deep Learning for Multimedia”. His current research interests include applications of data-driven models using machine learning particularly deep learning for audio-visual content understanding and distribution optimisation.

Posted in TEWI-Kolloquium | Leave a comment

Review: Estimating Space-Time Covariance from Finite Sample Sets [Slides]

The review of the TEWI colloquium of Dr. Stephan Alexander Weiss from May 22, 2019 comprises the slides (below):

Abstract:

Covariance matrices are central to many adaptive filtering and optimisation problems. In practice, they have to be estimated from a finite number of samples; on this, I will review some known results from spectrum estimation and multiple-input multiple-output communications systems, and how properties that are assumed to be inherent in covariance and power spectral densities can easily be lost in the estimation process. I will discuss new results on space-time covariance estimation, and how the estimation from finite sample sets will impact on factorisations such as the eigenvalue decomposition, which is often key to solving the introductory optimisation problems. The purpose of the presentation is to give you some insight into estimating statistics as well as to provide a glimpse on classical signal processing challenges such as the separation of sources from a mixture of signals.

Stephan Weiss. I am a Professor at the University of Strathclyde and head its Centre for Signal & Image Processing. My particular interests are adaptive filtering and array signal processing.

Posted in TEWI-Kolloquium | Leave a comment

Review: A Survey of Evaluation Techniques and Systems for Answer Set Programming [Video][Slides]

The review of the TEWI colloquium of Prof. Francesco Ricca from May 3, 2019 comprises the video and slides (below):

Abstract:

Answer set programming (ASP) is a prominent knowledge representation and reasoning paradigm that found both industrial and scientific applications. The success of ASP is due to the combination of two factors: a rich modeling language and the availability of efficient ASP implementations. In this talk we trace the history of ASP systems, describing the key evaluation techniques and their implementation in actual tools.

CV:

Francesco Ricca (www.mat.unical.it/ricca) is currently an Associate Professor at the Department of Mathematics and Computer Science of the University of Calabria, Italy. In the same Department he is Coordinator of the Computer Science Courses Council.
He received his Laurea Degree in Computer Science Engineering (2002) and a PhD in Computer Science and Mathematics (2006) from the University of Calabria, Italy, and received the Habilitation for Full Professor in Computer Science (INF/01) in 2017.
He is interested in declarative logic-based languages, consistent query answering, and rule-based reasoning on ontologies and in particular on the issues concerning their practical applications: system design and implementation, and development tools. He is co-author of more than 100 (peer-reviewed) publications including international research journals (30+), encyclopedia chapters, conference proceedings, and workshops of national and international importance. He has served in program committees of international conference and workshop, such as IJCAI, AAAI, KR, ICLP, LPNMR and JELIA, and has been reviewer for AIJ, JAIR, TPLP, JLC, etc. He is Area Editor of Association for Logic Programming newsletters, and member of the Executive Board of the Italian Association for Artificial Intelligence.

Posted in TEWI-Kolloquium | Leave a comment

Estimating Space-Time Covariance from Finite Sample Sets

Dr. Stephan Alexander Weiss | May 22, 2019 | 11:00 | B02.1.59

Abstract:

Covariance matrices are central to many adaptive filtering and optimisation problems. In practice, they have to be estimated from a finite number of samples; on this, I will review some known results from spectrum estimation and multiple-input multiple-output communications systems, and how properties that are assumed to be inherent in covariance and power spectral densities can easily be lost in the estimation process. I will discuss new results on space-time covariance estimation, and how the estimation from finite sample sets will impact on factorisations such as the eigenvalue decomposition, which is often key to solving the introductory optimisation problems. The purpose of the presentation is to give you some insight into estimating statistics as well as to provide a glimpse on classical signal processing challenges such as the separation of sources from a mixture of signals.

Stephan Weiss. I am a Professor at the University of Strathclyde and head its Centre for Signal & Image Processing. My particular interests are adaptive filtering and array signal processing.

Posted in TEWI-Kolloquium | Leave a comment

Why AI is shaping our games

Dr. Johanna Pirker | May 16, 2019 | 10:00 | B01.0.203

Abstract:

AI is used to create parts of our games. It provides intelligent enemy behavior, techniques such as pathfinding or can be used to generate in-game content procedurally. AI can also play our games. The idea to train computers to beat humans in game-like environments such as Jeopardy!, Chess, or soccer is not a new one. But can AI also design our games? The role of Artificial Intelligence in the game development process is constantly expanding. In this talk, Dr. Pirker will talk about the importance of AI in the past, the present, and especially the future of game development.

Bio:

Dr. Johanna Pirker is researcher at the Institute of Interactive Systems and Data Science at Graz University of Technology (TUG). She finished her Master’s Thesis during a research visit at Massachusetts Institute of Technology (MIT) working on collaborative virtual world environments. In 2017, she finished her doctoral dissertation in computer science on motivational environments under the supervision of Christian Gütl (TUG) and John Belcher (MIT). She specialized in games and environments that engage users to learn, train, and work together through motivating tasks. She has long-lasting experience in game design and development, as well as virtual world development and has worked in the video game industry at Electronic Arts. Her research interests include AI, data analysis, immersive environments (VR), games research, gamification strategies, HCI, e-learning, CSE, and IR. She has authored and presented numerous publications in her field and lectured at universities such as Harvard, Berlin Humboldt Universität, or the University of Göttingen. Johanna was listed on the Forbes 30 Under 30 list of science professionals.

Posted in TEWI-Kolloquium | Leave a comment

Use, Misuse, and Reuse of Continuous Integration Features

Prof. Shane McIntosh | May 2, 2019 | 14:00 | N.1.42

Abstract:

Continuous Integration (CI) is a popular practice where software systems are automatically compiled and tested as changes appear in the version control system of a project. Like other software artifacts, CI specifications, which describe the CI process, require maintenance effort. In this talk, I will describe the results of an empirical analysis of patterns of feature use and misuse in the Travis CI specifications of 9,312 open source systems. To help developers to detect and remove patterns of misuse, we propose Hansel and Gretel, anti-pattern detection and removal tools for Travis CI specifications. To help developers to rapidly develop and reuse common CI logic, we propose an extension to the TouchCORE modelling tool that allows users to select high-level features from CI feature models and generate an appropriate CI specification. To support this envisioned tool, we perform an initial analysis of common CI features using association rule mining, which yielded underwhelming results.

Bio:

Shane McIntosh is an assistant professor in the Department of Electrical and Computer Engineering at McGill University, where he leads the Software Repository Excavation and Build Engineering Labs (Software REBELs). He received his PhD in Computer Science from Queen’s University, for which he was awarded the Governor General of Canada’s Academic Gold Medal. In his research, Shane uses empirical software engineering techniques to study software build systems, release engineering, and software quality. More about his work is available online at http://rebels.ece.mcgill.ca/.

Posted in TEWI-Kolloquium | Leave a comment

A Survey of Evaluation Techniques and Systems for Answer Set Programming

Prof. Francesco Ricca | May 3, 2019 | 11:00 | S.1.42

Abstract:

Answer set programming (ASP) is a prominent knowledge representation and reasoning paradigm that found both industrial and scientific applications. The success of ASP is due to the combination of two factors: a rich modeling language and the availability of efficient ASP implementations. In this talk we trace the history of ASP systems, describing the key evaluation techniques and their implementation in actual tools.

CV:

Francesco Ricca (www.mat.unical.it/ricca) is currently an Associate Professor at the Department of Mathematics and Computer Science of the University of Calabria, Italy. In the same Department he is Coordinator of the Computer Science Courses Council.
He received his Laurea Degree in Computer Science Engineering (2002) and a PhD in Computer Science and Mathematics (2006) from the University of Calabria, Italy, and received the Habilitation for Full Professor in Computer Science (INF/01) in 2017.
He is interested in declarative logic-based languages, consistent query answering, and rule-based reasoning on ontologies and in particular on the issues concerning their practical applications: system design and implementation, and development tools.
He is co-author of more than 100 (peer-reviewed) publications including international research journals (30+), encyclopedia chapters, conference proceedings, and workshops of national and international importance. He has served in program committees of international conference and workshop, such as IJCAI, AAAI, KR, ICLP, LPNMR and JELIA, and has been reviewer for AIJ, JAIR, TPLP, JLC, etc. He is Area Editor of Association for Logic Programming newsletters, and member of the Executive Board of the Italian Association for Artificial Intelligence.

Posted in TEWI-Kolloquium | Leave a comment

Artificial Intelligence (AI) in media applications and services

Dr.-Ing. Christian Keimel | 9.5.2019 | 10:00 | S.1.42

Abstract: Artificial Intelligence (AI) is nowadays used frequently in many application domains. Although sometimes considered only as an afterthought in the public discussion compared to other domains such as health, transportation, and manufacturing, the media domain is also transformed by AI enabling new opportunities, from content creation e.g. “robojournalism” and individualised content to optimisation of the content production and distribution. Underlaying many of these new opportunities is the use of AI in its current reincarnation as deep learning for understanding the audio-visual content by extracting structured information from the unstructured data, the audio-visual content.

In this talk the current understanding and trends of AI will therefore be discussed, what can be done, what is done, and what challenges remain in the use of AI especially in the context of media applications and services. The talk is not so much focused on the details and fundamentals of deep learning, but rather on a practical perspective on how recent advances in this field can be utilised in use-cases in the media domain, especially with respect to audio-visual content and in the broadcasting domain.

Bio: Christian Keimel received his B.Sc and Dipl.-Ing.(Univ.) in information technology from the Technical University of Munich (TUM) in 2005 and 2007, respectively. In 2014 he received a Dr.-Ing. degree from TUM for his dissertation on the “Design of video quality metrics with multi-way data analysis.” Since 2013 he is with the Institut for Rundfunktechnik (IRT), the research and competence centre of the public service broadcasters of Austria, Germany, and Switzerland, where he leads the machine learning team, working on the applications of machine learning and AI in the broadcasting context. In addition, he is a lecturer at TUM for “Deep Learning for Multimedia”. His current research interests include applications of data-driven models using machine learning particularly deep learning for audio-visual content understanding and distribution optimisation.

Posted in TEWI-Kolloquium | Leave a comment

Towards 6DoF Adaptive Streaming Through Point Cloud Compression

Jeroen van der Hooft | 25.03.2019 | 16:00 | S.2.42

Abstract: The increasing popularity of head-mounted devices and 360-degree video cameras allows content providers to provide virtual reality video streaming over the Internet, using a relevant representation of the immersive content combined with traditional streaming techniques. While this approach allows the user to look around and move in three dimensions, the user’s location is fixed by the camera’s position within the scene. Recently, an increased interest has been shown for free movement within immersive scenes, referred to as six degrees of freedom (6DoF). One way to realize this, is by capturing one or multiple objects through a number of cameras positioned in different angles, creating a point cloud object which consists of the location and RGB color of a significant number of points in the three-dimensional space. While the concept of point clouds has been around for over two decades, it recently received increased attention by MPEG, issuing a call for proposals for point cloud compression. As a result, dynamic point cloud objects can now be compressed to bit rates in the order of 3 to 55 Mb/s, allowing feasible delivery over today’s mobile networks. In this talk, we use MPEG’s dataset to generate different scenes consisting of multiple point cloud objects, and propose a number of rate adaptation heuristics which use information on the user’s position and focus, the available bandwidth and the buffer status to decide upon the most appropriate quality representation of each of the considered objects. Through an extensive evaluation, we discuss the advantages and drawbacks of each solution. We argue that the optimal solution depends on the considered scene and camera path, which opens interesting possibilities for future work.

Bio: Jeroen van der Hooft obtained his M.Sc. degree in Computer Science Engineering from Ghent University, Belgium, in July 2014. In August of that year, he joined the Department of Information Technology at the same university, where he is currently active as a Ph.D. student. His main research interests include the end-to-end Quality of Experience optimization in adaptive video streaming, and low-latency delivery of immersive video content. During the first months of 2019, he worked as a visiting researcher in the Institute of Information Technology at the University of Klagenfurt, where he focused on rate adaptation for volumetric media streaming.

Web sitehttps://users.ugent.be/~jvdrhoof/

Posted in TEWI-Kolloquium | Leave a comment