The SaS Seminars are a permanent series of open seminars of the Division of Software and Systems (SaS) at the Department of Computer and Information Science (IDA), Linköping University. The objective of the seminars is to present outstanding research and ideas/problems relevant for SaS present and future activities. In particular, seminars cover the SaS research areas software engineering, programming environments, system software, embedded SW/HW systems, computer systems engineering, realtime systems, parallel and distributed computing, and theoretical computer science. -
Two kinds of seminars are planned:

• talks by invited speakers not affiliated with SaS,

• internal seminars presenting lab research to whole SaS (and other interested colleagues).

The speakers are expected to give a broad perspective of the presented research, adressing the audience with a general computer science background but possibly with no specific knowledge in the domain of the presented research. The normal length of a presentation is 60 minutes, including discussion.

The SaS seminars are coordinated by Christoph Kessler.

Previous SaS seminars (2021-2022)

Machine Learning for EDA and EDA for Machine Learning

Prof. Jiang Hu, Department of ECE, Texas A&M University, USA

Friday 7 Oct. 2022, 13:15, room John von Neumann, IDA

Abstract:
The wave of machine learning splashes to almost every corner of the world due to its unprecedented success. The first part of this talk will be focused on how to leverage machine learning for EDA (Electronic Design Automation). Specifically, a machine learning-based early routability prediction technique will be introduced. This technique provides a stochastic approach to handling non-deterministic data labels, which may exist in other machine learning applications. In the second part, an EDA technique for ML hardware acceleration will be presented. This is the first analytical approach to CNN hardware and dataflow co-optimization, and outperforms state-of-the-art methods in terms of both solution quality and computation runtime.

Speaker Bio:
Jiang Hu is a professor in the Department of Electrical and Computer Engineering at Texas A&M University. His research interests include design automation of VLSI circuits and systems, computer architecture optimization and hardware security. He has published over 240 technical papers. He received best paper awards at DAC 2001, ICCAD 2011, IEEE International Conference on Vehicular Electronics and Safety 2018 and MICRO 2021. He was the technical program chair and the general chair of the ACM International Symposium on Physical Design (ISPD) in 2011 and 2012, respectively. He was named an IEEE fellow in 2016.

Between Grepping and Deep-learning - Finding a Middle Ground in Software Anomaly Detection

Prof. Mika Mäntylä, M3S / ITEE / University of Oulu, Finland

Friday 7 Oct. 2022, 10:15, room John von Neumann, IDA

Abstract:
Understanding software behavior is particularly important for modern software development, where software must evolve quickly without sacrificing reliability. At the end of the day, the users do not care if poor software service is due to operations (Ops) or development (Dev) errors, thus, we must resolve both types of anomalies quickly. We can analyze software behavior using data from software testing or software operations meaning both fields can benefit from the same innovations. While state-of-the-art research, in this field, pushes towards ever more advanced but complex and computationally expensive deep-learning approaches, the state-of-the-practice still often uses grepping or diffing aided with domain knowledge. In this talk, I outline some solutions between the two extremes.

Speaker Bio:
Mika Mäntylä is a professor of Software Engineering at the University of Oulu, Finland. He received a D. Sc. degree in 2009 in Software Engineering from the Aalto University, Finland. His research interests include end of lifecycle software engineering activities like software testing, software maintenance, software operations. He is also interested in grey literature reviews, and behavioral and psychological aspects in software engineering. Currently, he applies mainly quantitative research methods from empirical software engineering, mining software repositories. He is keen to utilize natural language processing, classical machine learning, and deep learning to software engineering problems. He has previously worked as an assistant professor at the Aalto University, Finland, and as a post-doc at the Lund University, Sweden. His studies have appeared in journals such as IEEE Transaction on Software Engineering, IEEE Software, Empirical Software Engineering, and Information and Software Technology. He currently serves as an associated editor for IEEE Software. For more information, see mmantyla.github.io.

Scientific Machine Learning: Where We Are and Where We Need To Go

Dr. Chris Rackaukas, MIT

Friday, 10 June 2022, 10:15, room John von Neumann

Abstract:
Scientific machine learning (SciML) methods allow for the automatic discovery of mechanistic models by infusing neural network training into the simulation process. In this talk we will start by showcasing some of the ways that SciML is being used, from discovery of extrapolatory epidemic models, development of earthquake-safe buildings, to nonlinear mixed effects models in pharmacology. From there, we will discuss some of the increasingly advanced computational techniques behind the training process, focusing on the numerical issues involved in handling differentiation of highly stiff and chaotic systems. The viewers will leave with an understanding of how compiler techniques are being infused into the simulation stack to increasingly automate the process of developing mechanistic models.

Short bio:
Chris is a Research Affiliate in MIT CSAIL and previously an Applied Mathematics Instructor at MIT. Chris is the lead developer of the SciML Open Source Software Organization, which includes DifferentialEquations.jl solver suite along with hundreds of state-of-the-art packages for mixing machine learning into mechanistic modeling. Chris' work on high performance differential equation solving is the engine accelerating many applications from the MIT-CalTech CLiMA climate modeling initiative. As the Director of Scientific Research at Pumas-AI, Chris is the lead developer of Pumas, where Chris received the Emerging Scientist award from ISoP, the highest early career award in pharmacometrics. As the Director of Modeling and Simulation at Julia Computing, Chris is the lead developer of JuliaSim, where the work is credited for the 15,000x acceleration of NASA Launch Services simulations and recently demonstrated a 60x-570x acceleration over Modelica tools in HVAC simulation, earning Chris the US Air Force Artificial Intelligence Accelerator Scientific Excellence Award.

Switching the current - A megagame for improved understanding of energy systems

Dr. Ola Leifler, Department of Computer and Information Science, Linköping.

Thursday, 30 September 2021, 15:15-16:15, on zoom
(zoom link sent out by email to IDA; others who would like to attend please contact C. Kessler for the access data.)

Abstract:
This talk will present both a short background to and an overview of an upcoming cross-disciplinary research project with members from both several parts of LiU and two other university colleges. Originally, the project started as a pedagogical project that grew to become something much bigger. The goal of the project is to create a platform that enables increased, multi-perspectival understanding of energy systems, society, environment, and climate. Energy systems are complex, with a range of stakeholders and conflicting aims. A pedagogical challenge is to increase mutual understanding among relevant actors from a systems perspective. The project will develop a simulation-based megagame, conduct and study the outcome of several playtests in the coming years. The megagame simulates a region, its power relations, events, and values connected to the region's needs, production capabilities, and consequences for the environment and climate. The game's format is intended to provide a customizable platform that facilitates dialog and in-depth understanding of simulated events.

Short bio:
Ola Leifler is a senior lecturer at PELAB and shares his time between IDA and Didacticum. His main research interest is in understanding what can lead to radical shifts in our thinking so that we can contribute actively to the large-scale transformation that is needed for a sustainable society.

Previous SaS Seminars

Previous SaS Seminars

• 2020
• 2019
• 2018
• 2017
• 2016
• 2015
• 2014
• 2013
• 2012
• 2011
• 2010
• 2009
• 2008
• 2007
• 2006
• 2005
• 2004
• 2003
• 2002
• 2001

Page responsible: Christoph Kessler
Last updated: 2023-01-05