Master Thesis, 30 hp: Bayesian Theory of Mind

Master Thesis, 30 hp: Bayesian Theory of Mind

Are you a student eager to apply your theoretical knowledge and fresh perspectives to real-world challenges? At Saab, we believe that innovation thrives on new ideas, and your master thesis project could be the spark that ignites our next technological breakthrough.

Your role

We recognize the immense value that students bring to our company. Your academic rigor, combined with your enthusiasm for cutting-edge technology, allows you to approach problems with a unique and insightful lens. At Saab, you'll have the opportunity to collaborate with experienced engineers and specialists, gaining invaluable practical experience while making a tangible contribution to our growth and development.

Background

Autonomous systems operating alongside other decision-makers must often act with limited information and under time constraints. A racing vehicle may need to judge whether an opponent is aware of an attempted overtake, a drone may need to anticipate how another platform will react to incomplete sensor information, and a distributed autonomous system may need to coordinate without direct communication. In such settings, success depends not only on estimating the physical state of the world, but also on reasoning about what other agents know, believe, and are likely to do.

Current methods for so-called stochastic games in belief space can account for uncertainty in other agents but rely on strong simplifying assumptions to remain computationally tractable. In particular they avoid uncertainty over other agents’ beliefs and deeper belief hierarchies of the form “I think you think I think. . . ”. Addressing these limitations leads to a challenging problem at the intersection of Bayesian inference, probability theory, stochastic control, game theory, and numerical optimization.

Description of the Thesis Project
The central question is simple to state but difficult to answer: how much of another agent’s uncertainty and reasoning can an autonomous system model before the computational cost outweighs the strategic benefit? This MSc thesis project aims to investigate how richer Bayesian models of other agents can be represented and used for real-time strategic decision-making.

The project will build on recent work on stochastic dynamic games in belief space and study whether some of its simplifying assumptions can be relaxed without making the resulting algorithms computationally impractical. Possible directions include introducing higher-order beliefs, representing uncertainty about another player’s belief, and developing compact parametric, factored, or sampling-based approximations of these nested probability distributions.

The work will combine literature study, mathematical modelling, algorithm development, implementation, and numerical experiments, with the exact direction shaped by the students’ interests and findings. The project also has clear potential to develop into a conference paper if the work leads to a novel and computationally effective method for higher-order Bayesian reasoning in stochastic dynamic games.

 

Your profile

The project is suitable for 2 students in the final stage of Engineering Physics, Engineering Mathematics or equivalent and are about to begin a 30 hp thesis project.

You should be comfortable with probability theory, linear algebra, optimization, and programming. Courses in stochastic processes, Bayesian statistics or state estimation, optimal control, and numerical optimization are particularly relevant. A course in measure-theoretic probability or measure theory is considered a merit but is not a prerequisite.

We provide the support and guidance you need to translate your theoretical knowledge into practical solutions. Join us and become a driving force behind Saab's technological advancements!

This position requires that you pass a security vetting based on the current regulations around/of security protection. For positions requiring security clearance additional obligations on citizenship may apply.

Kindly observe that this is an ongoing recruitment process and that the position might be filled before the closing date of the advertisement.

Contact information

David Olsson, Manager

073-418 96 59

[email protected]

Mika Persson,Master Thesis Supervisor

073-434 11 93

[email protected]

What you will be a part of

Explore a wealth of possibilities. Take on challenges, create smart inventions, and grow beyond. This is a place for curious minds, brave pioneers, and everyone in between. Together, we achieve the extraordinary, each bringing our unique perspectives. Your part matters.

Saab is a leading defense and security company with an enduring purpose, to help nations keep their people and society safe. Empowered by its 28,000 talented people, Saab constantly pushes the boundaries of technology to create a safer and more sustainable world.

Saab designs, manufactures and maintains advanced systems in aeronautics, weapons, command and control, sensors and underwater systems. Saab is headquartered in Sweden. It has major operations all over the world and is part of the domestic defense capability of several nations. Read more about us here.