Tobias Morville

PhD, MSc in Economics


Trained economist with a PhD in Computational Neuroscience. My background is technical with a intent focus mathematical modeling. I have 5 years+ experience with designing, developing and delivering machine learning solutions in academia and industry.

Specialized in

Mathematical modeling, machine learning in production, data science as a process.

Tech stack

Everything Python, Git, Docker & Linux. Learning Rails as a hobby.


Nov 14 - Oct 17 Ph.D. in computational neuroscience, University of Copenhagen.

2010 - 2013 M.Sc. in Economics, University of Copenhagen.

2006 - 2010 B.Sc. in Economics, University of Copenhagen.

Key experience

Nov 17 - Senior Data Scientist,

As a Senior data scientist I have three main responsibilities: 1) Design and develop machine learning models for various industry cases. 2) Write the software that support those models in production. 3) Design and document procedures and workflow around data science as a process.

Machine learning

Trained a churn model (XGBoost) on time-series data from three major danish unions (millions of rows, thousands of features) and built a framework (Dash) for explaining model predictions per member based on Shapely additive values.

Deep learning

Built a fraud-detection model (classification of graph embeddings using TensorFlow and Keras) on a graph that contained every single danish company and its connections.

General development

Built a end-to-end workflow based on Docker containers which allows for full unit-testing of deployed models, complete reproduction of experiments and system agnostic model deployment.


Nov 17 - May 18 Research Consultant, Danish Research Centre for Magnetic Resonance.

As a research consultant I managed one major experiment, did end-to-end analysis of neuroimaging data and helped write a scientific article on theory derived from my thesis.


Design and execution of a fMRI experiment in which subjects were automatically fed small amounts of juice while interacting with a virtual game real time.

Image processing

Preprocessing and statistical analysis of behavioral and neuroimageing data (Python, Matlab).


Nov 14 - Oct 17 Ph.D. Fellow, University of Copenhagen.

My thesis is about how the brain computes reward. I developed a novel model for predicting dopaminergic activity, ran the experiment to test predictions and analysed the resulting data.

Theoretical development

Built a conceptual model of the reward system based on the free energy principle. Code up the circuit model of how this is implemented in the micro circuitry of the brain. The theory goes into detail with reinforcement learning, statistical physics and information theory.


Design and execution of several major neuroimaging experiments, often requiring building custom pieces of electronics and managing research assistants.

Image processing

Built several pipelines (Matlab, Python) for analysis of behavioral and neuroimaging data. In summary, a mix of computer vision, time-series statistics and machine learning.

Other experience

Oct 12 - Nov 13 Research Assistant, Danish Research Centre for Magnetic Resonance.

Aug 10 - Jan 14 Freelance consultant, Weekendavisen.

Apr 09 - Mar 12 Student Researcher, DAMVAD consultancy agency.

May 08 - Mar 09 Student Researcher, Wealth Management, SEB Enskilda.

Feb 07 - May 08 Student Researcher, Danish Oil Industry Association.


Research stays

Aug 15 - Feb 16 London Mathematical Laboratory, Dr. Ole Peters.

Nov 14 - Feb 16 The Francis Crick Institute, Professor Denis Burdakov.

Special training

Aug 16 Summer School on semi-supervised learning, DTU.

Jun 15 Complex Systems Summer School, Santa Fe Institute.

Feb 14 - Jul 14 TA, Neuroimaging Foundations, University of Copenhagen.

Dec 12 - Jul 13 External lecturer and supervisor, Metropolitan University College.


Peer reviewed

May 16 Tuning the Brake While Raising the Stake: Network Dynamics during Sequential Decision-Making. Journal of Neuroscience.

Feb 18 Patient profiling for success after weight loss surgery (GO Bypass Study): An interdisciplinary study protocol. Contemporary Clinical Trials.

In review

Jan 19 Neuro-computational Theories of Homeostatic Control. Physics of Life Reviews.

Sept 18 The Homeostatic Logic of Reward. Progress in Neurobiology.


July 17 Reward Signaling Under Glycemic Flux.

Conference abstracts

May 16 Decision-making and the Dynamics of Death. The 38th International Symposium of the Groupe de recherche sur le système nerveux central (GRSNC).

Nov 15 The Homeostatic Logic of Reward. Society for Neuroscience.

Jul 14 The Homeostatic Logic of Reinforcement Learning. Workshop on Information Theoretic Incentives for Artificial Life.


July 17 Rational, as on the savannah (Danish article). Weekendavisen #47 | Ideer | 22. Nov – 28. Nov 2013.


Nov 17 PhD Defense at University of Copenhagen. Homeostatic Choice Theory.

Jan 17 Neurograduate Symposium at University of Copenhagen. Homeostatic Control Theory.

May 16 The 38th Symposium of Neuroscience of Decision Making. Decision Making and the Dynamics of Death.

Nov 14 Society for Neuroscience. The Homeostatic Logic of Reward.

Jun 14 Conference for Artificial Life. The Homeostatic Logic of Reinforcement Learning.

Mar 13 CNEE Experimental Economics Workshop. Autocorrelation Biases in a Sequential Gambling Task.

May 13 CNEE Neuroeconomics Workshop. Homeostatic Reinforcement Learning.