Enhancing Semiconductor Scheduling with Maskable PPO and Genetic Algorithms

June 17th 2025 from 11.45 – 13.15 in outdoor HS Kosuta (V.1.27 in case of rain)

Univ.-Ass. Dott. Mag. Peyman Eftekhari
Department of Artificial Intelligence and Cybersecurity (AICS)

Title
„Enhancing Semiconductor Scheduling with Maskable PPO and Genetic Algorithms“

Short Abstract:
Scheduling in semiconductor manufacturing presents significant challenges due to vast action spaces and dynamic production constraints. My doctoral research aims to develop advanced optimization strategies by integrating Maskable Proximal Policy Optimization (PPO) with Genetic Algorithms (GA). Maskable PPO enables efficient navigation of large discrete action spaces by allowing only valid actions, thereby accelerating the learning process and improving policy outcomes. Simultaneously, Genetic Algorithms are utilized to refine scheduling heuristics and establish strong baseline solutions. In this talk, I will discuss the rationale for combining reinforcement learning and evolutionary approaches, highlight the current progress, and share preliminary experimental findings alongside future research directions.

Short Bio:
I am a PhD candidate at the University of Klagenfurt under the supervision of Professor Martin Gebser. Alongside my research, I work as a university assistant and currently teach Algorithms and Data Structures in the bachelor’s program. My research focuses on the intersection of reinforcement learning and evolutionary computation, with a particular emphasis on optimization challenges in semiconductor manufacturing scheduling. I specialize in applying Maskable PPO and Genetic Algorithms to solve large-scale, complex scheduling problems. My academic background includes computer science and artificial intelligence, with practical experience in machine learning, optimization techniques, and teaching.

Multi-UAV Route Planning and Task Assignment for Reforestation Seeding

May 13^th 2025 11.45 – 13.15 outdoor HS Kosuta (V.1.27 in case of rain)

Abstract of the talk:

Deforestation in many countries worldwide has led to a significant reduction
in forest areas and a rise in the frequency and intensity of fires, exacerbated by climate
change. The disappearance of native forests contributes to higher net carbon dioxide
emissions in the atmosphere, intensifying the greenhouse effect associated with global
warming [1]. A solution to that problem can be Using drones for aerial seeding in forest
restoration, which holds significant promise but encounters major obstacles that hinder
its efficiency and scalability. Existing methods typically involve blanket seeding across
the entire restoration area, resulting in a large number of seeds being dispersed, many of
which land in locations where they cannot successfully grow [2]. In this research, we
propose Multi-UAV route planning methods that focus on covering selected clusters of
forest land to seed those that have high germination rates instead of seeding all areas of
forest land.

CV:

Dipl.-Ing. Merna Tohfa is a research and teaching staff member at the Institute
of Networked and Embedded Systems at the University of Klagenfurt. Her research
focuses on routing planning techniques of multi-agent UAVs. She holds a master’s
degree in information and communications engineering, branch autonomous systems
and robotics from Klagenfurt University, Austria, and a bachelor of science degree in
engineering and materials science with a specialization in mechatronics engineering
from the German University in Cairo, Egypt.

Between Screen and Band saw line 4.0: Mediatization and Subjectivation in the Austrian Forestry and Sawmill Industry

April 30^th 11:45 – 13:15 HS Kosuta (S.2.05 in case of rain)

Manfred Rosenzopf, BA MA (Doctoral studies in Philosophy)

Abstract of the talk:

This presentation introduces a dissertation project that explores the mediatization and subjectivation of work in traditional industrial sectors. With the increasing integration of digital media and communication technologies, the Austrian forestry and sawmill industry is undergoing significant transformation. But what does this mean for the people working in this sector? The ongoing digitalization and technological transformation of work are of interest not only in media and communication studies but also in the sociology of work.

The project aims to analyse everyday experiences in the Austrian forestry and sawmill industry to examine how new digital work and communication practices influence work culture and social interactions. The dissertation is situated within the tradition of Workplace Studies, an interdisciplinary field concerned with the impact of new technologies on work and subjectivity. Theoretically, the project draws on the concept of communication as a form of agency and its effects on work and subjectivation in everyday life. Empirically, employees of a medium-sized forestry and sawmill company will be interviewed in qualitative studies, work processes will be observed, and documents will be analysed.

Preliminary research has shown that a discursive field around the topic of the digitalized world of work is emerging. The more deeply digital transformation penetrates the world of work, the more it alters its organization, conditions, routines, patterns of knowledge, and value systems. This will be examined and discussed using the Austrian forestry and sawmill industry as a case study.

Implicit Hate Speech – Non-Negative Stereotypes About Identity Groups

March 25th at 11:45 – 13:15 in S.1.05

Univ.-Ass. Tina Lommel, Bakk. MA (D!ARC / Computational Linguistics)

Abstract:

Hate speech is not always overtly offensive — it often appears in subtle, implicit forms. One specific example is seemingly neutral or even positive statements about identity groups that still reinforce stereotypes or convey discriminatory meanings (e.g., Women make good cooks). This presentation explores this form of implicit hate speech and introduces a newly developed dataset that systematically captures such statements. The impact of different linguistic formulations on the perception of offensiveness is examined, highlighting when seemingly harmless statements may be considered problematic. Additionally, the shifting nature of stereotypes depending on context is analyzed, illustrating how their perceived offensiveness varies. Finally, approaches to the automated detection of such linguistic patterns are discussed, along with the challenges involved.

Tina Lommel is a PhD candidate in Computational Linguistics and works at the Digital Age Research Center at the University of Klagenfurt. Her research focuses on improving the automatic detection of hate speech and implicit hate speech.