I am Adjunct Associate Professor in Computational Mathematics at the Department of Mathematical Sciences at Chalmers University of Technology and University of Gothenburg and Specialist in Engineering Mathematics at Saab.

My research concerns theory, applications and deep learning accelerated computational methods for statistics for SDE, stochastic control, Markov games and radar signal processing. The goal is to push the boundaries for what can be filtered, controlled, played and detected in real time. I work under the model-based machine learning paradigm, where strucure is kept to a maximal degree and learned models are introduced only where there is little structure, or for surrogate modelling to remove computational bottlenecks.

Here you find my Official Chalmers page, Google Scholar profile and LinkedIn profile.

E-mail: adam.andersson(at)chalmers.se

Image: Me and the Saab Giraffe 1X radar

Publications

Preprints and submitted articles

• M. Persson, J. Lidman, J. Ljungberg, S. Sandelius, A.A, Dynamic one-time delivery of critical data by small and sparse UAV swarms: a model problem for MARL scaling studies, 2025. PDF

• K. Bågmark, A. A., S. Larsson, Nonlinear filtering based on density approximation and deep BSDE prediction, 2025. PDF

• K. Bågmark, A. A., S. Larsson, F. Rydin, A convergent scheme for the Bayesian filtering problem based on the Fokker-Planck equation and deep splitting, 2024. PDF

• A. A. and F. Lindner, Poisson Malliavin calculus in Hilbert space with an application to SPDE, 2017. PDF

Journal articles

• K. Andersson, A. A., C. W. Oosterlee, The deep multi-FBSDE method: a robust deep learning method for coupled FBSDEs. Accepted for publication in J. Sci. Comput. (2026). PDF

• A. A., A. Lang, A. Petersson and L. Schroer, Finite element approximation of Lyapunov equations related to parabolic stochastic PDEs Appl. Math. Opt. (2025). PDF

• K. Bågmark, A. A., S. Larsson, An energy-based deep splitting method for the nonlinear filtering problem, Partial Differential Eq. Appl. (2023). PDF

• K. Andersson, A. A., C. W. Oosterlee, Convergence of a robust deep FBSDE method for stochastic control, SIAM J. Sci. Comput. (2023). PDF

• A. A., A. Jentzen and R. Kurniawan, Existence uniquness and regularity for stochastic evolution equations with irregular initial values, J. Math. Anal. and Appl. (2021). PDF

• A. A. and F. Lindner, Malliavin regularity and weak approximation of semilinear SPDE with Lévy noise, Discrete Continuous Dyn. Syst. Ser. B. (2019). PDF

• A. A., M. Hefter, A. Jentzen and R. Kurniawan, Regularity properties for solutions of infinite dimensional Kolmogorov equations in Hilbert spaces, Potential Anal. (2019). PDF

• A. A., A. Jentzen, R. Kurniawan and T. Welti, On the differentiability of solutions of stochastic evolution equations with respect to their initial values, Nonlinear Anal. (2017). PDF

• A. A. and R. Kruse, Mean-square convergence of the BDF2-Maruyama and backward Euler schemes for SDE with globally monotone coefficients, BIT Numer. Math. (2017). PDF

• A. A., M. Kovács and S. Larsson, Weak error analysis for semilinear stochastic Volterra equations with additive noise, J. Math. Anal. Appl. (2016). PDF

• A. A., R. Kruse and S. Larsson, Duality in refined Sobolev-Malliavin spaces and weak approximation of SPDE, J. SPDE Anal. Comp. (2016). PDF

• A. A. and S. Larsson, Weak convergence for a spatial approximation of the nonlinear stochastic heat equation, Math. Comp. (2016). PDF

Lecture notes

• Introduction to stochastic control and filtering, 2025. PDF

• Malliavin Calculus (unfinnished and incomplete but instructive), Chalmers, 2013. PDF

• Ornstein-Uhlenbeck theory in finite dimensions, Chalmers, 2012. Written under the supervision of P. Sjögren based on his lectures. PDF

• Brownian motion, Chalmers, 2012. PDF

• Markov processes, Chalmers, 2012. PDF

Supervision

Instructions for master thesis students

PhD students

• Kasper Bågmark is working on deep learning for the nonlinear filtering problem. The aim is to obtain approximate filters that scale better than particle filters. It is an academic project financed by the Wallenberg AI, Autonomous Systems and Software Program (WASP). Jointly supervised with Stig Larsson. I have defined the project and act main supervisor.

• Mika Persson is working on decentralized control of multi-function sensors. It is an industrial PhD position funded by Saab and WASP within the WASP Research Arena in Public Safety (WARA-PS). Jointly supervised with Axel Ringh and Ann-Brith Strömberg at Chalmers and Benjamin Svedung Wettervik at Saab. I have defined the project and am industrial supervisor.

• Karl Hammar is working on joint online system identification and state estimation for stochastic differential equations. It is an industrial PhD position funded by Saab and WASP. Jointly supervised with Lennart Svensson and Moritz Schauer at Chalmers and Benjamin Svedung Wettervik at Saab. Benjamin has defined the project and we are industrial supervisors.

Postdoc

• Per Ljung is working on radar detection in challenging heterogeneous and non-Gaussian clutter. It is an industrial postdoc position funded by Saab and WASP. Jointly supervised with Tomas McKelvey at Chalmers, Patrik Dammert at Chalmers and Saab, Mattias Ekström and Andreas Sundström at Saab. I have defined the project and am industrial supervisor.

Master thesis students

• Filip Berglund and Lukas Wenåker are working on linear quadratic games. Supervised together with Mika Persson.

Previous master thesis students

• Oliver Spjuth Large scale target tracking using transformers. Chalmers/Saab 2024. Oliver contiued as a target tracking analyst and AI developer at Saab.
• Alfred Wärnsäter and Viktor Nevelius Wernholm Efficient evaluation of target tracking using entropic optimal transport. Chalmers/Saab 2024. Awarded thesis of the year 2024 by the Swedish Operations Research Association (SOAF). Alfred continued as a PhD student in optimization at KTH and Viktor as a radar system engineer at Saab.
• Filip Rydin A deep learning method for nonlinear stochastic filtering. Chalmers 2024. Filip continued as a PhD student in AI and optimal control at Chalmers.
• Isak Wikman and Samuel Winqvist Prospects of neural SDE for Bayesian smoothing. Chalmers/Saab 2024. Isak continued as a radar system engineer at Saab and Samuel continued as junior game mathematician at Thunderkick.
• Emma Öijar Jansson Bayesian estimation of sea clutter parameters for radar, a stochastic approach. KTH/Saab 2023. Emma continued as a data scientist at ICA Sverige.
• Albin Ekelund Karlsson and Samuel Sandelius Adaptive radar illuminations with deep reinforcement learning, Chalmers/Saab 2023. Albin continued as a consultant at AFRY and Samuel as a radar simulator developer at Saab.
• Karl Hammar Fast Bayesian inference with piecewise deterministic Markov processes, Chalmers/Saab 2023. Karl continued as a radar system engineer at Saab and later as industrial PhD student at Saab/Chalmers/WASP.
• Mika Persson Towards deep learning accelerated sparse Bayesian frequency estimation, Lund University/Saab 2022. Mika continued as a radar systems engineer at Saab and later as industrial PhD student at Saab/Chalmers/WASP.
• Axel Nathansson Exploration of reinforcement learning in radar scheduling, Chalmers/Saab 2021. Axel continued as algorithm developer at Tobii.
• Elias Hölen Hannouch and Oskar Holmstedt Deep learning accelerated Bayesian inference for state space models, Chalmers/Smartr 2020. Elias and Oskar continued to work as developers at Ericsson and Oskar later as a PhD student in biomathematics at Gothenburg University.
• Anton Matsson and Victor Ohlsson Predicting customer behavior using adversarial imitation learning, Chalmers/Smartr 2020. Anton continued as a PhD student in AI at Chalmers (PhD Thesis) and Victor as software engineer at Kollmorgen.
• Klara Granbom On nonlinear machine learning methods for dose-response data in drug discovery, Chalmers/Smartr/IRLAB 2020. Klara continued as a developer at Collector Bank.
• Kristoffer Andersson Approximate stochastic control based on deep learning and forward backward stochastic differential equations, Chalmers/Syntronic 2018. Kristoffer continued as a PhD student in machine learning for mathematical finance at Centrum Wiskunde Informatica, Amsterdam (PhD Thesis).
• Gustaf Ehn and Hugo Werner Scalable reinforcement learning for a simulated production line, Lund University/Syntronic 2018. Gustaf continued as algorithm developer at Syntronic and Hugo as data scientist at Stena Line.
• Robin Andersson Sparse representation and image classification with the shearlet transform, Chalmers/Syntronic 2017. Robin continued as algorithm developer at Syntronic.
• Viktor Blomqvist and David Lidberg Swedish dialect classification using artificial neural networks and Gaussian mixture models, Chalmers/Syntronic 2017. Viktor continued as algorithm developer at Syntronic and David as machine learning engineer at RaySearch Laborarories.
• Leander Schroer Numerical approximation of operator Riccati equations for distributed control of SPDE, TU-Berlin 2016. Leander continued as a consultant at Sopra Steria Consulting.
• Yueleng Wang Efficient computation of the strong and weak error for linear SDE, Chalmers 2014. Yueleng continued to study a second master in statistics at the University of Windsor.

Master thesis students supervised by close collegues within joint projects

• Benjamin Elm Jonsson and Bekir Fazlija Stochastic differential games and decentralized control. Saab/Chalmers 2025. Benjamin continued as a PhD student in mathematics at Humbold Univerität zu Berlin and Bekir as a radar system engineer at Saab.

• Jacob Ljungberg Communication Relaying Networks Using Autonomous Drones: Solving Cooperative Markov Games Using Multi-Agent Reinforcement Learning. Saab/Chalmers 2025. Jacob continued as a radar system engineer at Saab.

• Gustav Birath Blom and Isak Nilsson Learning neural SDEs for Bayesian filtering and smoothing. Saab/Chalmers 2025. Gustav and Isak continued as a radar system engineer at Saab.

• David Ahnlund and Axel Nilsson Diffusion models for generative modelling of a posteriori probability measures in target tracking. Saab/KTH 2025. David and Axel continued as system engineers at Saab.

Presentations on YouTube

• Lecture for the appointment to associate professor 2020: Riccati and Lyapunov equations for control and weak approximation of stochastic PDE, Chalmers.

• Gothenburg Artificial Intelligence Alliance 2020 conference: Removing computational bottlenecks with deep learning.

• Gothenburg Artificial Intelligence Alliance MeetUp 2018: Deep learning for equation solving and technical computations.

Bio

Positions

• 2023-ongoing: Saab, Specialist in Engineering Mathematics.
• 2021-ongoing: Chalmers University of Technology, Adjunct Associate Professor.
• 2020-2022: Saab, Radar Systems Engineer.
• 2019-2020: Smartr, Chief Scientist and consultant.
• 2016-2019: Syntronic, Team Leader and consultant.
• 2015-2016: TU Berlin, Postdoctoral researcher.
• 2009-2015: Chalmers University of Technology, PhD student.

Degrees

• 2021: Chalmers University of Technology, Associate Professor (Oavlönad Docent) in Computational Mathematics.
• 2015: Chalmers University of Technology, PhD in Mathematics. Thesis
• 2009: Chalmers University of Technology, MSc in Engineering Mathematics. Thesis
• 2006: Chalmers University of Technology, BSc in Automation and Mechatronics.

Recreation

• Älv, Bååth & Andersson. A film by my paddling friend Jacob Kasprup Haagensen about a trip in the Vistas river in 2020. Shown at the Nordic Adventure Film Festival in Copenhagen 2022.

• Kayaking in the Byske river 2019. A film by my brother Arvid Bååth.