Tutorial and State of the Art Course in topical areas related to TECoSA
Below are descriptions of the modules to be offered in the first round of the course. Attendance is free to all TECoSA members. To register, please email email@example.com stating which module(s) you wish to take. Alternatively, calendar invitations will be issued to everyone on the tecosa-all mailing list. You can register by accepting the invitations that are of interest to you. (If the module runs on more than one date, please be sure to accept all the relevant invitations.)
5G Security Standardization and Cloud Security Challenges, insights from Ericsson (key areas: Security)
Format: 3-hour workshop
Location: on-line (via Teams)
Timing: Friday 29 October 2021, kl.13-16
Module led by: Ericsson: Prajwol Kumar Nakarmi (firstname.lastname@example.org) and János Kövér (email@example.com)
Content: Part 1: A sneak peek into how 3GPP – the defacto standardization organization for all Gs – works in general. Then, we will describe the most important security and privacy features that were introduced in 5G, making 5G the most trustworthy mobile network generation ever. Part 2: To highlight problems and identify research opportunities: *Introduction (Threat model in cloud, ideal wanted position); *Solution with TEEs (Trusted Execution Environment) and short-term challenges; *Potential residual threats (e.g. side channel attacks, trust in HW vendors).
Trustworthiness and Dependability in Edge-based CPS (key areas: Safety, Security, Predictability)
Format: Two half-day workshops, plus preparation and homework
Location: KTH Campus, with “join on-line” option
Timing: Thurs 24 Feb kl.9-12 CET and Thurs 3 March kl.9-12 CET
Module led by: KTH: Martin Törngren (firstname.lastname@example.org)
Content: An overview of established views and definitions; relationships between concepts; trends; “schools”. Flipped class-room sessions (with preparation + assignment); discussion; collaborative effort to create a web/wiki page with “TECoSA” key concepts.
Abstract: As Cyber-Physical Systems (CPSs) become smarter, more automated, connected and collaborating, their new capabilities provide unprecedented opportunities for innovation in a large number of industrial and societal domains. As the same time, future CPS will incorporate a computing continuum, from smart devices over the edge to the cloud, often forming part of systems of systems. As such future CPS are adopted (in industry, society and infrastructures), we will even more than now be relying on their availability and correct functioning. This implies that various failures, intrusions, and non-intended features, misuse, and negative emergent behavior may have very drastic effects on our lives. Trustworthiness and dependability thus become key requirements for future CPS.
Machine Reasoning Explainability (an AAMAS tutorial) (key areas: Predictability)
Format: One 3.5-hour tutorial
Location: Online (via Teams)
Timing: Weds 9 March kl.9-12.30 CET
Module lead by: Ericsson: Kristijonas Cyras, Swarup Kumar Mohalik, Alexandros Nikou, Alessandro Previti (with support from Anusha Pradeep Mujumdar and Aneta Vulgarakis Feljan (email@example.com)
Content: *Concept; *Technical Part 1(Inference-Based Explanations; SAT/SMT Analysis of Inconsistencies; Argumentation-Based Explanations); *Technical Part 2 (Explainable Planning; Symbolic RL, Explainability; Causal Approaches to Explainability); *Conclusions.
Abstract: As a field of AI, Machine Reasoning (MR) uses largely symbolic means to formalize and emulate abstract reasoning. Studies in early MR have notably started inquiries into Explainable AI (XAI) – arguably one of the biggest concerns today for the AI community. Work on explainable MR as well as on MR approaches to explainability in other areas of AI has continued ever since. It is especially potent in modern MR branches such as argumentation, constraint and logic programming, and planning. We hereby aim to provide a selective overview of MR explainability techniques and studies in hopes that insights from this long track of research will complement well the current XAI landscape.
Industrial, Reliable Wireless Networking – Fundamentals and Systems (key areas: Predictability)
Format: Two 1.5-hour workshops
Location: Online via Zoom with hybrid option (KTH Campus) if situation allows
Timing: Thurs 17 March kl.16-17.30 CET and Fri 18 March kl.16-17.30 CET
Module led by: KTH: James Gross (firstname.lastname@example.org)
Content: *Underlying principles; *Communication theory background; *Implementation principles; *Examples (iWLAN, IO-Link, 5G, 802.11be). The content is split into two parts: “fundamentals” and “systems”. For the fundamentals, we will cover (a) Application layer requirements (b) latency/reliability challenges in wireless systems (c) theoretical results over the last decade (finite-block length models, medium access schemes, scheduling). For the systems, we will cover (a) Wifi Evolution; (b) Cellular Systems; (c) Other approaches.
Abstract: Wireless systems have been commercialized over the last thirty years mostly as systems to foster communication among humans. Applications like telephony, client-server apps and lately also video services are good examples of this footprint. However, over the last decade, other use cases have become more and more the focus of the research community, addressing essentially requirements that originate from industrial automation scenarios. These scenarios call for much lower latencies while simultaneously emphasizing the reliability of transmissions. In this module, we cover the most important aspects of this development, both from a research perspective (covering fundamentals) as well as from as from a systems perspective, considering the performance of contemporary standards as well as the next wave of systems to come.
ML Security and Privacy (key areas: Security)
Format: Two 2-hour seminars, plus individual study and homework
Location: KTH Campus
Timing: Mon 4 April kl.10-12 CET and Mon 25 April kl.10-12 CET
Module led by: KTH: György Dán (email@example.com) and Raksha Ramakrishna (firstname.lastname@example.org)
Content: *Taxonomy of attacks on ML; *Adversarial attacks on classification; *ML fingerprinting and watermarking; *Definitions of privacy; *Approaches to privacy preserving learning.
Fundamentals of Bayesian Inference using Probabilistic Programming (key areas: Safety)
Format: Two 3-hour seminars plus preparation and homework
Location: KTH Campus with “join on-line” option. Please register at: https://www.kth.se/form/6254058494395f32cdb8aaed
Timing: Weds 1 June kl.14-17 and Weds 7 June kl.14-17 (all CET)
Module led by: KTH: David Broman (email@example.com)
Content: *Basics of Bayesian inference (SMC, MCMC); *Bayes rule, probabilistic modeling; *Using probabilistic programming to solve Bayesian probabilistic problems. Given through interactive lectures, hands-on exercises and take home exercises.
Python for Data Science and Machine Learning (key areas: Predictability)
Format: Four 1-hour workshops plus self-study
NEW Timing: Weds 17 Aug kl.15-16, Weds 24 Aug kl.15-16, Weds 31 Aug kl.15-16, and Weds 7 Sept kl.15-16 (all CET)
Module led by: AFRY: Markus Skogsmo (firstname.lastname@example.org)
Content: Learn how to use Python as a tool for Data Science and Machine Learning purposes. This course is aimed at beginners and can serve as a first introduction to coding in Python – which begins with the fundamentals of Python and ends with some basic examples of Machine Learning predictions. This course was appreciated internally at AFRY and served for some novice users as a stepping stone, to use Python as a tool in their work as well as for their interests.
Format: Three 2-hour workshops
Location: Kista Campus, with “join on-line” option
Timing: Weds 5 Oct kl.13-15, Weds 12 Oct kl.13-15 and Weds 19 Oct kl.13-15 (all CET)
Module led by: KTH: Elena Dubrova (email@example.com)
Content: *Introduction to cryptography (secret and public cryptosystems, encryption and authentication basics); *Attack vectors; *Anti-tamper techniques.