Fredrik Sandin

Contact

To get in contact please book a meeting, connect on LinkedIn, or phone for urgent matters. I prefer dialogue, efficient co-creation meetings and focused working sessions over emails (motivation here).

Oral re-exams and Q&A about education can also be scheduled via the book-a-meeting link above.

Research

The LTU Machine Learning group has about thirty members working on basic research projects and application/challenge-driven projects. I'm particularly interested in neuromorphic technologies and lead projects and courses in that area, including projects where neuromorphic design, machine learning, computational physics, and physical instantiation of information processing are used to solve real-world interaction problems. This include projects focusing on sensor/detector and intelligent systems co-design where constraints (energy, power, latency, dynamic range, online learning, etc) challenge conventional and digital approaches. We have experience of mixed-signal neuromorphic circuits, algorithms and systems, as well as numerous machine learning projects involving industrial data, problems and collaboration. See Google scholar for further information.

Projects:

• Upcoming, 2026-2030: Picosecond-scale Photonic Heterogeneous Integrated Neuromorphic Detector (PHINDER), Pathfinder Open, under grant agreement preparation. Coordinating with Lama Alkhaled.
• Shaping the Future with Neuromorphic Technology (NEUTEC): Towards Global Leadership in Next-Generation Intelligent Systems, October 2025-March 2026. Coordinating with Sabine Mayer.
• Neuromorphic Innovation Platform Sweden (Vinnova, 2024-02857), with KTH, Lund University, Uppsala University, FOI, ABB, Ericsson, and SAAB. Coordinated with Sabine Mayer.
  • Report: Neuromorphic Sweden: Status, Needs and Recommendations (2025)
• Testbed for machine monitoring with ultra-low-power neuromorpic sensing and intelligence (ÅForsk, ref no 24-530, 1.8 MSEK), with Pär Marklund and Daniel Strömbergsson.
• Differential optimization of neuromorphic-digital intelligent systems (Jubileumsfonden och Kempestiftelserna, JCSMKJF23-0003, 2 MSEK), coordinator, with TVM and INFN in Padova.
• Fast, Robust, and Efficient NanoPhotonic Neuromorphic Computing (WASP-WISE preproject, LiU-2023-00139), with Anders Mikkelsen, NanoLund.
• The Kempe Foundations NCS (dnr SMK-1429 and JCK-1809, 4.4 MSEK). PhD thesis (2023): Event-Driven Architectures for Heterogeneous Neuromorphic Computing Systems.
• Hybrid Digital-Neuromorphic Computing Services: Interoperability, Interfacing and Programming (Vinnova, 2023-01363), with RISE.
• Neuromorphic postdoc projects, funded by Kempestiftelserna (JCSMK23-0218) and Creaternity.
• KnowIT Fast (PiiA, dnr 2019-02533, 4.6 MSEK). Coordinator. PhD thesis completed June 2025:Technical Language Supervision and AI Agents for Condition Monitoring.
• ECSEL JU Arrowhead Tools (no 826452, 84 MEUR); EU ITEA3 AutoDC (dnr 2018-02232, 58 MSEK).
• VALD (Vinnova FFI, dnr 2021-05035, 3.4 MSEK); ALDEE (Vinnova FFI, dnr 2019-03073, 3.5 MSEK); MetMaskin (Vinnova, dnr 2018-02666, 4.7 MSEK).
• STINT Institutional Grant (dnr IG2011-2025, 1.5 MSEK), neuromorphic technologies for wireless sensor networks (active during the years 2012-2017), coordinated with Sven Molin.

PhD students:

• Kenneth Paulsen, neuromorphic TinyML (co-supervising, with Hui Han).
• Awais Khan, materials for neuromorphic computing (co-supervising, with Tommaso Dorigo).
• Rickard Brännvall (employed at RISE), privacy preserving and scalable machine learning at the edge, started in AutoDC.
• Karl Löwenmark (2025), Technical Language Supervision and AI Agents for Condition Monitoring (thesis, GPTech agent and tools).
• Saleha Javed (2025), Cloud-based IoT and Collaborative Learning for Cyber-Physical System of Systems (thesis, co-supervisor, with Marcus Liwicki)
• Carl Borngrund (2024), Towards Deep-learning-based Autonomous Navigation in the Short-loading Cycle (thesis, co-supervisor, with Ulf Bodin).
• Gustav Grund Pihlgren (2023), Deep Perceptual Loss and Similarity (thesis, co-supervisor, with Marcus Liwicki).
• Mattias Nilsson (2023), Event-Driven Architectures for Heterogeneous Neuromorphic Computing Systems (thesis, with Foteini Simistira Liwicki).
• Muhammad Ahmer (2023), Intelligent fault diagnosis and predictive maintenance for a bearing ring grinder, (thesis, co-supervisor, industrial PhD at SKF, with Pär Marklund).
• Jacob Nilsson (2022), Machine Learning Concepts for Service Data Interoperability (thesis, worked in Arrowhead Tools, with Jerker Delsing and Marcus Liwicki).
• Kim Albertsson (2021), Machine Learning in High-Energy Physics: Displaced Event Detection and Developments in ROOT/TMVA (thesis, CERN funded, with Andreas Hoecker).
• Siddharth Dadhich (2018), Automation of Wheel-Loaders (thesis, co-supervisor, with Ulf Bodin).
• Sergio Martin del Campo Barraza (2017), Unsupervised feature learning applied to condition monitoring (thesis, SKF University Technology Center on Advanced Condition Monitoring, with Jerker Delsing).
• Blerim Emruli (2014), Ubiquitous Cognitive Computing: A Vector Symbolic Approach (thesis, with Jerker Delsing).

Postdocs:

• Neuromorphic computing postdoc grant (Kempe Foundations), recruiting.
• Daniel Strömbergsson (2025), Co-design of neuromorphic condition monitoring sensor system.
• Sergio Martin del Campo Barraza (now adjunct senior lecturer in our group), unsupervised machine learning for automation of wind turbine condition monitoring system, two years funded by SKF (2018-2019).

We are members of WASP and ELLIS since 2022 and MODE since 2023.

Courses

• Neuromorphic Computing (D7064E), deepening course focusing on neuromorphic technologies and spiking neural networks with applications and project work.
  • Most projects are done in collaboration with industry or resarch groups/projects.
  • Also available as a 3rd cycle course including an individual project for Ph.D. students.
    
• Neural Networks and Learning Machines (D7046E), advanced level course on artificial and spiking neural networks, entry point for D7064E (above) and Advanced Deep Learning (D7047E).
  • Also available as a 3rd cycle course including an individual project for Ph.D. students.
• Programming for Scientific Computing (D7066E), advanced level course on programming and software engineering for scientific and high-performance computing (HPC).
• Master thesis courses in Engineerng Physics and Electrical Engineering (X7010E, X7011E, X0003E).

I lecture also in D0036E Programming for Machine Learning, D0032 Introduction to AI, and D0028E Programming and Digitalization. Before I was examiner for E0003E Electric circuit theory, D0011E Digital design, D0017E Introduction to programming for engineers, and D0036E. Before 2010 I taught courses in physics, including mechanics, thermodynamics, waves, optics, quantum physics, and 3rd-cycle lectures in astrophysics, cosmology and quantum field theory at finite temperature.

We strive for active learning and promoting efficient communication and learning-centric use of tools including AI. We rely on micromodules and video-recorded (short/focused) lectures for presenting theory, methods and tools to avoid repetitive one-directional "communication". We mostly use oral examination, including computer labs and open-ended projects that are orally examined.

Since 2014, I coordinate the "Teknisk fysik och elektroteknik" program at LTU (Civilingenjör, 300Hp) in collaboration with Andreas Almqvist.

Background

With a scholarship from the Natural Science Research Council I did an MSc diploma work in ATLAS at CERN (2001). I have a PhD in Physics (2007, LTU) and was one of the first students in the Swedish Graduate School of Space Technology. My thesis focused on dense states of matter in neutron stars. I received a “New-Talents” award for an original work in theoretical physics at the International School of Subnuclear Physics in Erice for work in collaboration with Johan Hansson. First postdoc funded by FNRS and focused on computational and fundamental physics at IFPA in Belgium (2008-2009).

A growing interest for the physics of brains and development of neuromorphic technologies, inspired by visits to leading groups in Heidelberg (Karlheinz Meier) and Zurich (Giacomo Indiveri and Elisabetta Chicca), made me shift research focus. Much thanks to Lennart Gustafsson and his course SME063 Artificial Neural Networks (which I studied as an undergrad in 2001) I got an opportunity for a second postdoc in brain-like computing (2010-2011) at EISLAB with Jerker Delsing, where I became assistant professor (until April 2016), associate professor (until March 2021) and presently work as professor in the Machine Learning group with Marcus Liwicki. I'm particularly interested in the functional and structural principles of (small) brains and the development of neuromorphic technologies for next-generation intelligent systems, including an interest for the physics/emergence of phenomenal consciousness as a potential route towards more potent AI objectives and safeguarding concepts.

Support from The Kempe Foundations have been key for initiating the neuromorphic activities at LTU over the last decade. In particular the 2014 Gunnar Öquist Fellowship including 3 MSEK for research and mentorship by Gunnar was a key enabler. At that time I also served as a technical committee member of the SKF–LTU University Technology Center, as a research theme leader (2013-2018) in the Intelligent Industrial Processes area of excellence, and as a member of the Applied AI Innovation Hub. Over the years I also served in several faculty workgroups focusing on education/program development.

I served as reviewer for top journals (like TNNLS, Neuromorphic Computing and Engineering, Physical Review, Frontiers in Neuroscience, Physical Review) and the European Commission; reviewer of fundamental science applications, e.g., for the Australian and Swedish Research Councils; reviewer of professor applications; reviewer of PhD theses; workshop organizer; program chair. I coordinated national and EU research proposals. Professional education include: Organisational health and leadership (2014), Previa AB; Intellectual property (2014), SKF European Patent Office; Docent qualification (2012), LTU; Media training (2011), Kalix Folkhögskola; Teaching and learning in higher education (2010), LTU; Project management (2009), Astrakan AB; Personal leadership (2006), Personal Management International (PMI).

Some highlights

• Neuromorphic Sweden: Status, Needs and Recommendations (2025).
• Technical Language Supervision and AI Agents for Condition Monitoring (2025), PhD thesis and GPTech agent with tools for industrial fault diagnosis by Karl Löwenmark (proposal written with Stephan Schnabel at SKF in 2018, long before LLMs and agent frameworks became popular).
• Event-Driven Architectures for Heterogeneous Neuromorphic Computing Systems (2023), first PhD thesis at LTU focusing on neuromorphic computing. Mattias is postdoc at Zenke Lab.
• Machine Learning Concepts for Service Data Interoperability (2022), PhD thesis in Arrowhead investigating machine-learning based dynamic interoperability.
• Gunnar Öquist Fellowship Award and 3 MSEK grant from The Kempe Foundations.
• Ubiquitous Cognitive Computing: A Vector Symbolic Approach (2014), first PhD student and thesis I had the honour and pleasure to supervise. Blerim worked on neurosymbolic and "hyperdimensional" computing and is now senior lecturer at Lund University. His pioneering work inspired Evgeny and Denis at LTU to pick up work on hyperdimensional computing.
• Dark matter in neutron stars hypothesis with Paolo Ciarcelluti: Paper, paper, impact. We showed that hidden-sector dark matter candidates can modify the structure and observations of neutron stars.
• ISSP award for an Original Work in Theoretical Physics (note who signed the award: Prof. 't Hooft and Prof. Zichichi; the work by 't Hooft and his mentor Martinus Veltman has been considered as the most advanced work awarded a Nobel Prize in physics (anecdote: at another wonderful PhD summer school in Nijmegen 2003, I asked 't Hooft about his view on the physics of consciousness and was unable to fully comprehend the answer :o).
• Preon star hypothesis. Paper, paper, observational predictions, Nature news, Phys. Rev. Focus, New Scientist.
  

Older artefacts

• 3FCS code and web interface. This framework was developed and used for color superconducting quark matter calculations in several papers about neutron stars with quark matter cores and related phase diagrams, including this highly cited paper in Phys. Rev. D.
• Femtolensing code and web interface. Calculates the gravitational lensing signatures of low-mass (10^14 -10^17 kg) compact interstellar objects. This code was developed when working on the preon star hypothesis, see highlights above.
• N-dimensional random projection code, link. Early work on representation learning for cognitive computing based on hyperdimensional vectors (toddler embeddings).
• CBVS code. More early work on cognitive computing, related publications here and here.
• Dircheck. A portable/interoperable tool for verification of the integrity of file archives across operating systems and storage media formats.
• Python code for distributed computing over ssh, developed around 2005 during PhD studies.
• ASOUND Matlab plugin, developed to enable asynchronous background processing of sound.
• Online collaborative writing in Latex, link. Developed and used with some collaborators before the era of sharelatex, overleaf etc.
• ProdUp, sawmill production analysis, optimization and planning application developed during highschool later acquired by Norrskog Wood Products.