Management & Society

The project "Game Over Eva(sion): Securing Deep Learning with Game Theory" aims to protect deep learning classifiers against targeted attacks. Deep learning classifiers are not only popular in scientific research but are also more and more adapted to the daily life, as self-driving cars, smartphones, and digital personal assistants use these kind of algorithms. Unfortunately, recent research has shown that almost all deep learning classifiers are vulnerable to so-called evasion attacks. In an evasion attack, an attacker can slightly modify a benign object and by this achieves a misclassification with a very high probability. The robustness of deep learning classifiers to these attacks is an open problem. We will model the competition of an attacker who is able to launch an evasion attack and the defender who wants to train a deep learning classifier that is robust against such an attack with means of game theory. In a first step, we will analyze which concepts of other research areas, such as adversarial classification, can be translated to the domain of secure deep learning. Then, we will develop a game-theoretic model that captures all relevant aspects and dependencies between the attacker's and the defender's strategies. In the course of the project, we expect the first theoretically well-founded results on the achievable security of deep learning classifiers in the presence of evasion attacks. Furthermore, we will evaluate existing countermeasures against evasion attacks to gain insights about their optimality when facing a strategic attacker. Finally, we want to implement the key properties from out theoretical models in a practical deep learning classifier. We will compare our classifier against other state-of-the-art classifiers in terms of robustness against evasion attacks and accuracy on benign input objects. By this, we can validate if it is possible to make deep learning classifiers robust against evasion attacks. The expected results of the project will enable us to judge if deep learning classifiers are suitable for scenarios where an attacker has incentives to explicitly fool them, i.e., security-critical areas and widespread consumer products.

Industry and business are currently undergoing a significant transformation towards an automated and digitalized world. The challenges for manufacturing companies are manifold and include technical, economic, and sociological aspects. The Tiroler Rohrer GmbH is currently facing the challenge of carrying out a wall thickness optimization on one of the gravity die casting machines (machine 4), whereby the main focus will be on the tip end. The following reasons are given: - wall thickness at the tip end is still difficult to control - The simulation did not yield a clear result for control - peak wall thickness depends on too many parameters - Through automated measurement of the wall thickness profile for each pipe, new process data is available - By reading out all (available) process parameters, patterns between process parameters and wall thicknesses should be made visible.

The Lindner company manufactures tractors for green, mountain and arable farming and vehicles for municipal services (Unitrac), which are already equipped with various sensors. These sensors are used to collect different vehicle data, which are stored on a server. From this data memory the data can be tapped and prepared for further processing. A uniform software structure, evaluation tools, apps and a user interface are missing for the analysis of the data. The company Lindner would like to achieve an optimization of the work process based on the analysis of the vehicle data, coupled with external data (e.g. weather forecast). This includes, among other things, automated driving, ploughing, snow removal with the help of GPS data, irrigation of plants (in viticulture), calculation of the required amount of road salt based on the weather situation, calculation of the actual consumption of road salt, etc. In winter 2018/19, special attention will be paid to the further data collection and processing of the actual amount of road salt in municipal applications. Thus, with the support of weather forecasts and GPS data, the amount of road salt should be predicted and optimized.

Integration of various topics in the context of a changing world of work: * women‘s careers in mathematics and computer science * creation & agility of jobs, lifelong learning, talented diversity & innovativeness, innovative technologies, organizational design * ESCO european skills and competences framework Women in Technology - Education & Career