PhD Topics at the
Graduate School
of Artificial Intelligence

Shape Cutting-Edge Research!

Research at the Graduate School is closely linked to research conducted at the supervisor's institutes. Learn more about the exciting topics we offer. 

Deep Learning at the Institute for Machine Learning

Sepp Hochreiter, head of the Institute for Machine Learning at the JKU, is globally renowned for his work on the vanishing gradient and his Long Short-Term Memory (LSTM) networks. These are considered important cornerstones of what has later become known as Deep Learning.


Research topics at the institute cover many fields of modern Deep Learning techniques including, among others, perception for self-driving cars, reinforcement learning, generative adverserial networks and toxicology prediction. Perception is an essential part of a self-driving vehicle and is being explored by developing new recurrent neural network architectures. Reinforcement learning models have vastly gained momentum over the last few years. At the institute, actor-critic systems are refined by applying LSTM-based reward distribution models. Generative adversarial network architectures are studied under the prospect of classical and quantum mechanical field theories. Many chemical compounds (drug candidates) fail in the late phases of the drug development pipeline. Therefore, at the institute Deep Learning and biclustering methods are being developed to identify new on-target and off-target effects in various drug design projects.

Potential Ph.D. Topics:

  • Reinforcement Learning
  • Drug discovery, target prediction, and toxicoloy
  • Deep Learning for self-driving cars
  • Deep Learning for generative models

Logical Reasoning at the Institute of Formal Models and Verification (FMV)

Core competencies at the Institute of FMV lie in the foundations of computational logic. We apply logic to develop efficient reasoning engines, such as our award-winning SAT and SMT solvers. Using our tools, we strive to solve hard constraint satisfaction problems such as planning and scheduling problems. We apply logic for reasoning pertaining to the correctness of software and hardware, for example in approaches based on bounded model-checking, a widely-used verification technique that was co-invented by Armin Biere.

Potential Ph.D. Topics:

  • Reasoning with SAT/SMT/QBF
  • Development of efficient reasoning techniques
  • Combination of Formal Methods and Machine Learning

Acoustic Perception via Deep Learning at the Institute of Computational Perception (CP)

CP, headed by Gerhard Widmer, is one of the world's leading research labs for computational music and sound perception, defining the state-of-the-art in several fields and winning numerous international scientific competitions (see the CP Awards Page). By combining and further developing methods from signal processing, pattern recognition, probabilistic modelling, and (deep) machine learning, we develop algorithms and models designed to solve tasks ranging from music perception and real-time music tracking, to expressive music performance, to general audio recognition and sound event detection.

Potential Ph.D. Topics:

  • Expectation-guided perception in music and/or general sound classification and event detection
  • Adversarial approaches to sound perception and transformation
  • Acoustic perception from massive data with incomplete label information

Computer Algebra and Symbolic Computation at the Institute for Algebra

Modern mathematics depend heavily on the use of computers. Using algebraic computations, it is possible to now prove theorems that go far beyond the capabilities of traditional pen-and-pencil reasoning. For over 30 years, mathematicians in Linz have played a major role in developing the underlying technology and they continue to be highly active at the forefront of research in this area. Our academic activities cover the entire research development chain, ranging from rather theoretical computability considerations and the design and analysis of efficient algorithms to implementation considerations and applications.

Potential Ph.D. Topics:

  • Develop algorithms to prove and find mathematical formulas
  • Use machine learning to find more efficient algorithms for computer algebra
  • Use computer algebra to find more efficient algorithms for machine learning

Engineering through Machine Learning at the Institute for Software Systems Engineering (ISSE)

ISSE, under the management of Alexander Egyed, is one of the world's leading software and systems engineering institutes (e.g., Microsoft ranked ISSE 16th worldwide). Research at the institute covers a wide area of software engineering from requirements to capture software, systems architecture, design and testing, to maintenance. Engineering is an inherently creative process that requires rigorous attention to details. However, engineering is also a collaborative, human centric process with adhoc activities. Engineering automations are few and rare – not just during programming but also during modeling, testing or maintenance. This PhD topic explores uses for machine learning in context of software engineering. On the most basic level, we envision the application of machine learning for reasoning with incomplete, uncertain, and/or incorrect software engineering knowledge. Justifiably, most engineering tools follow the philosophy of “garbage in-garbage out”. How could a tool reason correctly in the presence of errors? How could a tool reason at all if the input is incomplete? Yet, we must. Much of engineering is about uncertainty, incompleteness, and incorrectness. If our tools are only useful once we have complete and correct information, then arguably these tools are not useful during much of the engineering process.

Potential Ph.D. Topics:

  • Predicting the Impact of Software and Systems Changes onto various Engineering Artifacts (i.e., what is affected when a requirement changes?)
  • Understanding Traceability from Requirements, Model, to Code
  • Handling the Exponential Features Combinations of Software and System Variants
  • Self-Optimization, Self-Adaptation, and Self-Healing of Partially and/or Fully Autonomous Systems (e.g., Smart Factory or Smart City)

Learn More

Pervasive Computing at the Institute of Pervasive Computing

Alois Ferscha, the head of the Institute of Pervasive Computing, recently created the COMET K1 Center Pro2Future to focus on “Cognitive Industrial Systems” which enables the application of AI (general AI) methods in industry (products and product systems). The institute is particularly interested in “Aware Systems” which utilizes a previous project's machine-learning methods for situation recognition, activity detection, work process identification, vital and emotional state recognition, as well as recognition of attention and intentions of individuals. The goal is to equip machines with human abilities such as perception, understanding, interpretation, memory, learning, reasoning, which should lead to cognitive autonomous policies and actions in the environment. Basic research is conducted in machine perception, cognitive robotics, cognitive decision systems as well as cognitive products.

For more information, contact our team members directly.