A Special Session on ‘Video Coding for Large Scale HTTP Adaptive Streaming Deployments‘ was organized by Christian Timmerer (Alpen-Adria-Universität Klagenfurt, Austria), Mohammad Ghanbari (University of Essex, UK), and Alex Giladi (Comcast, USA) on July 2 at the 35th Picture Coding Symposium (PCS) 2021. Read more about it here.
With his master thesis about “Animating Characters using Deep Learning based Pose Estimation”, Fabian Schober won the “Dynatrace Outstanding IT-Thesis Award” (DO*IT*TA). The award brings attention to extraordinary theses, motivates creativity, and provides insight into modern technologies.
In his thesis, Fabian Schober focuses on animating 2D (video game) characters using the PoseNet pose estimation model. He delivers a proof of concept on how new machine learning technologies can assist in video game development. Read more at the University press release (German only).
Successful review of the first research phase: Christian Doppler ‘pilot’ laboratory ATHENA to transition to a regular CD laboratory two years after launch. Read more about it at ATHENA website and the press release at AAU website.
Co-chairs:
TF7: Immersive Media Experiences (IMEx)
Immersive media applications are entering our daily lives starting from VR/AR/360° video applications to multi-sensory/multimedia experiences potentially addressing all human senses rather than focusing on hearing and seeing. The overall goal of providing Immersive Media Experiences (IMEx) to end-users is giving them the sensation of being part of the particular media which shall result in a worthwhile, informative user and quality of experience.
The actual objectives of this task force are as follows:
Our project „ADAPT“ started in March 2021, during the most critical phase of the COVID-19 outbreak in Europe. The demand for Personal Protective Equipment (PPE) from each country’s health care system has surpassed national stock amounts by far.
Learn more about it in an interview with Univ.-Prof. DI Dr. Radu Aurel Prodan in University Klagenfurt´s journal „ad astra“ (pdf).
The presentation has been accepted to the main-track of the Austrian-Slovenian HPC Meeting (ASHPC’21). Meeting will be organized in a hybrid format on 31 May – 2 June, 2021 at the Institute of Information Science in Maribor, Slovenia.
Title: Automated Workflows Scheduling via Two-Phase Event-based MILP Heuristic for MRCPSP Problem
Authors: Vladislav Kashansky, Gleb Radchenko, Radu Prodan, Anatoliy Zabrovskiy and Prateek Agrawal
Abstract: In today’s reality massive amounts of data-intensive tasks are managed by utilizing a large number of heterogeneous computing and storage elements interconnected through high-speed communication networks. However, one issue that still requires research effort is to enable effcient workflows scheduling in such complex environments.
As the scale of the system grows and the workloads become more heterogeneous in the inner structure and the arrival patterns, scheduling problem becomes exponentially harder, requiring problem-specifc heuristics. Many techniques evolved to tackle this problem, including, but not limited to Heterogeneous Earliest Finish Time (HEFT), The Dynamic Scaling Consolidation Scheduling (DSCS), Partitioned Balanced Time Scheduling (PBTS), Deadline Constrained Critical Path (DCCP) and Partition Problem-based Dynamic Provisioning Scheduling (PPDPS). In this talk, we will discuss the two-phase heuristic for makespan-optimized assignment of tasks and computing machines on large-scale computing systems, consisting of matching phase with subsequent event-based MILP method for schedule generation. We evaluated the scalability of the heuristic using the Constraint Integer Programing (SCIP) solver with various configurations based on data sets, provided by the MACS framework. Preliminary results show that the model provides near-optimal assignments and schedules for workflows composed of up to 100 tasks with complex task I/O interactions and demonstrates variable sensitivity with respect to the scale of workflows and resource limitation policies imposed.
Keywords: HPC Schedule Generation, MRCPSP Problem, Workflows Scheduling, Two-Phase Heuristic
Acknowledgement: This work has received funding from the EC-funded project H2020 FETHPC ASPIDE (Agreement #801091)
ADAPT started with the online Kickoff meeting, coordinated by Prof. Radu Prodan.
Prof. Radu Prodan is a keynote speaker at Memphis DATA 2021, 25th-26th March 2021.
Talk Abstract: We live in a digital world estimated to host around 4 billion Internet users and 10 billion of mobile connections generating 2.5 billion billion of data every day. Managing and extracting value from this sheer amount of raw data requires deep software analysis tools on massive distributed and parallel computing infrastructures aggregating billions of cores and threads. The talk gives an overview of the research activities at the University of Klagenfurt, Austria, on optimising system software support for extreme-scale data processing applications, with focus on scientific simulations, social media and massively multiplayer online games.
Title: WELFake: Word Embedding over Linguistic Features for Fake News Detection
Authors: Pawan Kumar Verma (Lovely Professional University, India | GLA University, India), Prateek Agrawal (University of Klagenfurt, Austria | Lovely Professional University, India), Ivone Amorin (MOG Technologies | University of Porto, Portugal), Radu Prodan (University of Klagenfurt, Austria)
Abstract: Social media is a popular medium for dissemination of real-time news all over the world. Easy and quick information proliferation is one of the reasons for its popularity. An extensive number of users with different age groups, gender and societal beliefs are engaged in social media websites. Despite these favorable aspects, a significant disadvantage comes in the form of fake news, as people usually read and share information without caring about its genuineness. Therefore, it is imperative to research methods for the authentication of news. To address this issue, this paper proposes a two phase benchmark model named WELFake based on word embedding (WE) over linguistic features for fake news detection using machine learning classification. The first phase pre-processes the dataset and validates the veracity of news content by using linguistic features. The second phase merges the linguistic feature sets with WE and applies voting classification. To validate its approach, this paper also carefully designs a novel WELFake dataset with approximately 72,000 articles, which incorporates different datasets to generate an unbiased classification output. Experimental results show that the WELFake model categorises the news in real and fake with a 96.73% which improves the overall accuracy by 1.31% compared to BERT and 4.25% compared to CNN models. Our frequency-based and focused analyzing writing patterns model outperforms predictive-based related works implemented using the Word2vec WE method by up to 1.73%.
Acknowledgement: ARTICONF project
The full paper has been accepted to the main-track of the International Conference on Computational Science (ICCS’21). Conference will be organized in a virtual format on 16-18 June, 2021.
Title: Monte-Carlo Approach to the Computational Capacities Analysis of the Computing Continuum
Authors: Vladislav Kashansky, Gleb Radchenko, Radu Prodan
Abstract: This article proposes an approach to the problem of computational capacities analysis of the computing continuum via theoretical framework of equilibrium phase-transitions and numerical simulations. We introduce the concept of phase transitions in computing continuum and show how this phenomena can be explored in the context of workflow makespan, which we treat as an order parameter. We simulate the behavior of the computational network in the equilibrium regime within the framework of the XY-model defined over complex agent network with Barabasi-Albert topology. More specifically, we define Hamiltonian over complex network topology and sample the resulting spin-orientation distribution with the Metropolis-Hastings technique. The key aspect of the paper is derivation of the bandwidth matrix, as the emergent effect of the “low-level” collective spin interaction. This allows us to study the first order approximation to the makespan of the “high-level” system-wide workflow model in the presence of data-flow anisotropy and phase transitions of the bandwidth matrix controlled by the means of “noise regime” parameter. For this purpose, we have built a simulation engine in Python 3.6. Simulation results confirm existence of the phase transition, revealing complex transformations in the computational abilities of the agents. Notable feature is that bandwidth distribution undergoes a critical transition from single to multi-mode case. Our simulations generally open new perspectives for reproducible comparative performance analysis of the novel and classic scheduling algorithms.
Keywords: Complex Networks, Computing Continuum, Phase Transitions, Computational Model, MCMC, Metropolis-Hastings, XY-model, Equilibrium Model
Acknowledgement: This work has received funding from the EC-funded project H2020 FETHPC ASPIDE (Agreement #801091)