This is a personal page, for which KTH requires a disclaimer in Swedish. With that out of the way, hello, my name is Magnus Boman, welcome, and I do research on

Artificial Intelligence

and how it

I am a professor at KTH since 2003 in Intelligent Software Services. I am affiliated with:

My application area has chiefly been health for the last 15+ years, namely computational epidemiology (digital epidemiology), adaptive treatments in psychiatry, and precision medicine, the latter under the lead of Anna Martling at KI (2021-). I am also leading an AI strategy project for the President of KI (2020-). In addition, Region Stockholm is funding my work at MedTechLabs (2023-).

Because of the recent interest in federated learning, I have made my Ph D thesis from 1993 on this topic available again, thanks to the Internet Archive I was able to locate a .ps file from my Web page mid-90s so a PDF can now be found here.

I am also

Jump down for more on:


PhD students

I am currently supervising where Jan is at Stockholm University, and I supervise him through my habilitation degree there. I also co-supervise the following PhD students, where Nils and Pontus are at KI. I have had the pleasure of supervising the following people, sorted by their graduation date.

Current courses

Learning Machines (ID2225/IK3616), where ID2225 is ending in 2024, teacher, course responsible, examiner

Research Methodology and Scientific Writing (II2202), teacher, assistant, responsible chiefly for Year 1 students in the Data Science programme

Critical Perspectives on Engineering and Construction of ICT Systems (IK3615), examiner

Philosophy of Science (IC2002), teacher, course responsible, examiner

Philosophy of Science (FIV3000), teacher, course responsible, examiner

Research Methodology in Computer Science (FID3023), teacher, course responsible, examiner


M Sc theses

I always have a stimulating number of master students, which means too many. But feel free to ask about opportunities nevertheless. Please do not expect any of the links to theses below to work, but most theses are available in the open portal DIVA. They are listed in pretty much reverse chronological order (most recent ones on top) and the list is not exhaustive. For some, I was the examiner and for others the supervisor, but I was actively involved with all of the below theses.

  • Pierre Meyrat: "Spatiotemporal PET reconstruction with Learned Registration" (KTH)

  • Glacier Ali and Rebecka Gustavsson: "Risk Stratification of Acute Coronary Syndrome using Machine Learning: An analysis of CLEOS-CPDS data" (KTH B Sc thesis, Receiver: Danderyd hospital, Stockholm)

  • Rosario Silva Sepulveda: "Using Multimodal Machine Learning for Analysing Multifactorial Causes of Disease: The case of childhood overweight and obesity in Mexico" (KI)

  • Núria Rodríguez-Ruiz: "Multimodal Data Interpretation to Support Clinical Decision-Making: A Conceptual Design for Molecular Tumour Board Meetings for Lymphoma" (KI)

  • Added Kina: Task-Agnostic Knowledge Distillation of mBERT to Swedish" (KTH)

  • Ahmad Sadek: "Primary stage Lung Cancer Prediction with Natural Language Processing-Based Machine Learning" (KTH, Receiver: KI)

  • David Mellin: "Improving a Few-shot Named Entity Recognition Model Using Data Augmentation" (KTH, Receiver: Findwise)

  • Styliani Katsarou: "Improving Multilingual Models for the Swedish Language - Exploring Cross-Lingual Transferability and Stereotypical Biases" (KTH, Receiver: Peltarion)

  • Sevket Melih Zenciroglu: "Comparing Non-Bayesian Uncertainty Evaluation Methods in Chromosome Classification by Using Deep Neural Networks" (KTH, Receiver: Arkus AI)

  • Sarthak Langde: "QPLaBSE: Quantized and Pruned Language-Agnostic BERT Sentence Embedding Model - Production-ready compression for multilingual transformers" (KTH, Receiver: Sinch)

  • Robin de Groot: "Developing a Mobile Eye Tracking Solution using Transfer Learning" (KTH, Receiver: Lexplore)

  • Juan Aldamis Orcajo: "A Machine Learning Approach to the analysis of mortality in patients with cardiovascular diseases" (KTH, Receiver: Intelligent Data Analysis Laboratory, Valencia Uni)

  • Omar Emilio Contreras Zaragoza: "Explainable Antibiotics Prescriptions in NLP with Transformer Models" (KTH, Receiver: Peltarion)

  • Boris Bubla: "DistilLaBSE: Task-agnostic distillation of multilingual sentence embeddings" (KTH, Receiver: Sinch)

  • Bazil Muzaffar Kotriwala: "Predictive Maintenance of Construction Equipment using Log Data" (KTH: Receiver: Volvo Construction Equipment)

  • Einar Lennelöv: "Overcoming generative likelihood bias for voxel-based out-of-distribution detection" (KTH, Receiver: RaySearch)

  • Annika Gerigoorian and Maha Kloub: "A Cross-Sectional Technology Acceptance Study of an AI CAD System in a Breast Screening Unit" (KTH B Sc thesis, Receiver: Sankt Görans hospital, Stockholm)

  • Tengfei Ma: "A Graph Attention plus Reinforcement Learning Method for Antenna Tilt Optimization" (KTH, Receiver: Ericsson)

  • Esoon Ko: "Product Matching through Multimodal Image and Text Combined Similarity Matching" (KTH: Receiver: APPRL)

  • Johan Vikström: "Comparing decentralized learning to Federated Learning when training Deep Neural Networks under churn" (KTH)

  • Mastafa Foufa: "Anomaly Detection Across Multiple Languages" (KTH, Receiver: Microsoft, Dublin)

  • Pietro Alovisi: "Static Branch Prediction with Representation Learning" (KTH)

  • Eric Samikwa: "Multi-modal IoT sensing with edge computing for early warning systems" (KTH, Receiver: RISE, Kista)

  • Xin Ren: "Machine Learning for Genome-Wide Association Study Data" (KTH, Receiver: KI, Huddinge)

  • Marcos Fernández Carbonell: "Automated Multimodal Emotion Recognition" (KTH, Receiver: Stockholm University, Psychology Dept)

  • Ronja Jösch: "Managing Microservices with a Service Mesh---An implementation of a service mesh with Kubernetes and Istio" (KTH, Receiver: Sveriges Television)

  • Tongtong Fang: "Learning from noisy labels by importance reweighting: A deep learning approach" (KTH, Receiver: RIKEN AIP, Tokyo)

  • Christian Boersmaa: "Designing a platform to communicate posture and movement data to medical professionals" (KTH, Receiver: Qinematic)

  • Ottar Gudmunsson: "Gating Networks in Learning Machines for Multimodal Data: Decision Fusion on Single Modality Classifiers" (KTH, Receiver: RISE)

  • Diego Roa: "Analysis of Short Text Classification Strategies using Out-of-Domain Vocabularies" (KTH, Receiver: Celonis)

  • Hadi Sotudeh: "Predicting Undesired Business Process Behavior using Supervised Machine Learning" (KTH, Receiver: Celonis)

  • Mattias Arro: "Label-Efficient Multi-Objective Machine Learning for e-Commerce" (KTH, Receiver: SICS)

  • Erik Droh: "T-Distributed Stochastic Neighbor Embedding Data Preprocessing Impact on Image Classification using Deep Convolutional Neural Networks" (KTH B Sc thesis)

  • Staffan Jansson Aldenfalk: "Prediction of Alarms in a Pump Station using Neural Networks" (KTH, Receiver: Xylem Inc.)

  • Tijmen Verhulsdonck: "One Shot Object Detection for Tracking Purposes" (KTH, Receiver: ETH / UC Berkeley)

  • Daniyal Shahrokhian: "Syna: Emotion Recognition based on Spatio-Temporal Machine Learning" (KTH, Receiver: SICS)

  • Akash Singh: "Anomaly Detection for Temporal Data using Long Short-Term Memory (LSTM)" (KTH, Receiver: Ericsson)

  • Filip Stojanovski: "Churn Prediction using Sequential Activity Patterns in an On-Demand Music Streaming Service" (KTH, Receiver: Spotify)

  • Dominik Harz: "Trust and verifiable computation for smart contracts in permissionless blockchains" (KTH, Receiver: SICS)

  • Mazen Aly: "Automated Bid Adjustments in Search Engine Advertising" (KTH, Receiver: Precis Digital)

  • Andrea Azzini: "Deep Learning Quantification: Extracting Quantitative Information from Images using Convolutional Autoencoders" (KTH)

  • Henny Selig: "Continuous Event Log Extraction for Process Mining" (KTH, Receiver: Signavio, Berlin)

  • Robert Velu: "BarcodeCheck: Retrieval of Product Information" (KTH, Receiver: SICS)

  • Ali Merikh: "A Triplestore built on Document-Oriented Databases" (KTH, Receiver: SICS)

  • Mikael Stoltz: "A Parrondo asset allocation strategy based on a discrete time microstructure model" (KTH, Receiver: SICS)

  • Abdulah Kayal: "A Neural Networks Filtering Mechanism For Foreign Exchange Trading Signals" (KTH, Receiver: SICS)

  • Stephanie Habib: "Web Caching in IPTV Portals" (KTH, Receiver: Ericsson)

  • Anusharani Gopu: "Using non-medical risk factors related to dementia and cognitive decline for developing an evidencebased e-health tool" (KTH, Receiver: SICS)

  • Khurram Shairyar Qalander: "Translating policy statements of Abu Dhabi health authorities into Business Rules" (KTH)

  • Eric-Oluf Svee: "Time Geography and Structuration Theory as a Basis for Human-centered Microsimulation" (KTH, Receiver: SICS)

  • Mathias Engvall: "The design of an order flow simulator to validate algorithmic trading strategies" (KTH, Receiver: OMX)

  • Sharenya Krishnan: "Text-Based Information Retrieval Using Relevance Feedback" (KTH, Receiver: SICS)

  • Jon Gretar Gudjonsson & Gary MacRitchie: "Strategy Adapting Trading Agent" (KTH, Receiver: SICS)

  • Henrik Aronsson: "Modeling strategies using predictive analytics: Forecasting future sales and churn management" (KTH, Receiver: Attollo)

  • Maryamossadat Nematollahi Mahani: "Resolving ambiguities in service robots' behavior" (KTH)

  • Alexander Bea: "Re-engineering two financial trading invoicing systems into one integrated solution" (KTH, Receiver: Nordicstation)

  • Somshree Mukherjee: "Ranking System for IoT Industry Platform" (KTH, Receiver: SICS)

  • Baki Cakici: "Pickman's Machine: A Reasoning Architecture" (KTH, Receiver: SICS)

  • Katarina Gyllenbäck: "Narrative bridging: A specification of a modelling method for game design" (KTH, Receiver: SICS)

  • Nicolas Grasset: "Multi-agent microsimulations with time geography" (KTH, Receiver: SICS)

  • Khandaker Tabin Hasan: "Metadata-rich Web development methodology" (KTH, Receiver: Univ of Trento)

  • German Gonzalez: "Kinematic tracking and activity recognition using motion primitives" (KTH, Receiver: Brown Univ, NY)

  • Ka Lok Chan: "Integration of browsing and searching for semantic information spaces" (KTH)

  • Zinat Sultana: "Improving the Design Method of the Analysis Pipeline for Spatial Transcriptomics" (KTH, Receiver: SciLifeLab/KI)

  • Robert Nilsson "Genetiska algoritmer och neurala nät tillämpade på aktiehandel" (KTH, Receiver: SICS)

  • Asim Ghaffar: "Visualization of Epidemic Spread" (KTH, Receiver: SMI)

  • Dimitrios Konstantinou & Sofia Savvidou "Benchmark för mobiltelefoner" (KTH, Receiver: SICS)

  • Niklas Boström: "Implementing advanced business logic for stock trade agents" (SU, Receiver: SICS)

  • Daniel Hilmersson: "SmartTrader: Implementing an Advanced Business Logic Agent for Agent Trade Servers" (Mid-Sweden University, Examiner: Prof Love Ekenberg, Receiver: SICS)

  • Tobias Hasselrot: "Fair Bandwidth Allocation in Internet Access Gateways: Using Agent-Based Electronic Markets" (KTH, Receiver: SICS)

  • Kristina Andréasson: "Modelling Infectious Disease Control" (SU, Receiver: SICS)

  • Jesper Johansson & Michael Poijes: "Agent Shell for Stock Market Systems" (SU, Receiver: SICS)

  • Jan Odelstad: "Artificial Agents and Norms" (SU)

  • Peter Hultman: "A Risk-Averse TAC Agent" (KTH, Receiver: SICS)

  • Wah-Sui Almberg: "Improved Pricing on the Stock Market with Trading Agents" (SU, Receiver: OM)

  • Vincent Hauser & Joakim Orrhult: "Network Monitoring for Networking Agents" (SU, Receiver: SICS)

  • Rikard Wigforss: "Artificial Intelligence for Real-Time Computer Games" (KTH, Receiver: Grin)

  • Terese Lund: "Cell-phones and Context" (KTH, NADA, IPLab, Examiner: Anna Swartling, Receiver: SICS)

  • Björn Qvarsell: "Programming SIM Cards for Mobile Services" (KTH, Receiver: Razorfish)

  • Hilde Hasli & Jing Cheng: "Smart Downloading of Multimedia" (SU, Receiver: Ericsson Radio Systems)

  • Jari Pyyluoma: "The origin of law - An Epistemological Artificial Societies Approach" (SU)

  • Conny Loung: "Agent Behaviour in Groups" (KTH, Receiver: Ducere)

  • Magnus Hübner: "Pagent - A Search Engine Interface for Mobile Terminals" (KTH, Receiver: Toshiba, Tokyo branch)

  • Martin Kalén: "A JINI-Based Multi-Agent System" (KTH, Receiver: Spectral)

  • Erik Welander: "Decision Analysis on the Internet - The DELTA/CGI Pronouncer" (SU)

  • Sven-Erik Ceedigh: "Trading Agents" (SU)

  • Göran Eneroth: "Smart Equipment" (KTH)

  • Jannilla Lidén: "Agent Related Properties of Dutch and English Auctions" (SU)

  • Christian Guttmann: "A Software Architecture for Four-Legged Robots" (KTH)

  • David Lybäck: "Transient Diversity in Multi-Agent Systems" (KTH)

  • Olle Källander: Behaviour-Based Bots in Quake2" (KTH)

  • Håkan Younes: "Current Tools for Assisting Intelligent Agents in Real-Time Decision Making" (KTH)

  • Marcus Hammarberg & Mats Bovin: "Decision Support for Software Agents" (SU, Receiver: Telia Research)

  • Per Thorsson-Börd: "Adding Resource-Specific Information to Agents in a Multi-Agent System" (KTH, Receiver: Telia Research)

  • Sebastian Wand: "Building a Monitor for a Multi-Agent System" (KTH, Receiver: Telia Research)

  • Helena Åberg: "Agent Roles in RoboCup Teams" (KTH)

  • Åsa Åhman: "Decision Control in RoboCup Teams" (KTH)

  • Henrik Bergström: "Applications, Minimisation, and Visualisation of Finite State Machines" (SU)

  • Henrik Engström & Johan Kummeneje: "DR ABBility: Agent Technology and Process Control" (SU, Receiver: ABB Network Partner)

  • Daniel Ejderberg: "Creating a User Agent in a Distributed Information Retrieval Environment" (KTH, Receiver: Telia Research)

  • Ioannis Tzikas: "Agents and Objects: The Differences" (SU)

  • Lars Rasmusson: "Socially Controlled Global Agent Systems" (KTH, Receiver: SICS)

  • Andreas Rasmusson: "Interactive Security Assistance for End-User Supervision of Untrusted Programs" (KTH, Receiver: SICS)

  • Gabriel Fuchs: "On Diminishing the Gap Between Speed and Correctness in a Decision-Making System" (SU)


    Select publications

    ResearchGate, Google Scholar, and other profiles are open and hold (mostly meaningless) quantitative data. The below list can be roughly categorised into the following five areas.

    Computational Health

    Data Analytics

    Networked Foresight and Management Science

    Foundational AI and Multi-Agent Systems, plus Applications outside of Health

    Logic, Mathematics, and Physics


    Creative Commons License
    This work is licensed under a Creative Commons Attribution-ShareAlike 2.5 Sweden License.

    Most recent edit by m a b @ k t h . s e on 27 May 2023.