Master Thesis, 30 HP: Enhancement of Deep Learning Results Interim Result Data
Machine learning has found increased usefulness in multides of aspects of analytical work. By leveraging the technique, it is usually possible to bypass elaborate calculations with machine learning algorithms. When the background theory in physics may not be fully known, it is also sometimes possible to go directly from observed data to new predictions.
Description of the master thesis
This thesis would aim to explore if machine learning results can be enhanced by using interim results as guidance for the algorithm. For example, using time series data for a moving object to determine a final state instead of only having av input state and an output state. Different techniques that should be of interest are LSTM, reinforcement learning and GANs.
This Master Thesis is suitable for 1-3 students with interest in machine learning and/or physics.
You are at the end of your computer sciences or physics education and about to start your Master Thesis work for 30 HP.
What you will be a part of
The Explosive Technology department consists of 20 employees engaged in work within development, testing and analysis in a variety of areas such as of explosives, pyrotechnics, chemistry and life-time technology. The section also performs calculations regarding thermochemistry and the effect of explosives. The employees work with many different products, which means a great variety of tasks.
Fredrik Dahlin, Manager
+46 73 4463163
Victor Björkgren, Master Thesis Supervisor
+46 73 4461032
If you aspire to help create and innovate whilst developing yourself in a challenging team setting, Saab may well have the perfect conditions for you to grow. We pride ourselves on a nurturing environment, where everyone is different yet we share the same goal – to help protect people.