Logo: to the web site of Uppsala University

uu.sePublications from Uppsala University
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Bridging Spiking Neural Networks and Deep State-Space Models
Uppsala University, Disciplinary Domain of Science and Technology, Technology, Department of Electrical Engineering, Signals and Systems.ORCID iD: 0009-0007-9645-3328
2026 (English)Doctoral thesis, comprehensive summary (Other academic)
Description
Abstract [en]

Modern artificial intelligence achieves impressive results, but at a high energy cost. Neuromorphic computing seeks to reduce this cost by using brain-inspired principles. Central to this approach are spiking neural networks (SNNs), in which neurons communicate through sparse binary events called spikes. Separately, in signal processing and control theory, state-space models (SSMs) have long offered a systematic description of dynamical systems. More recently, SSMs have re-emerged as powerful deep learning models for processing long sequences, called deep SSMs. Spiking neurons are themselves dynamical systems, yet there has been limited interaction between SNNs and deep SSMs. This thesis brings these two together in a common framework and uses this common framework to explore two research directions.

The first direction asks how neuron models should be designed to capture richer temporal dynamics while remaining computationally efficient. Drawing on tools from SSMs, we propose enrichments along four axes: (i) We generalise neuron models with a single input and a single output channel to multiple input and multiple output channels, widening the bandwidth constrained by binary spikes. (ii) We generalise the reset mechanism, and show that it can stabilise learning even when the underlying linear dynamics are unstable. (iii) We equip neurons with structured, learnable delays that provide direct access to past inputs. (iv) We allow neurons' dynamics to vary over time, enabling them to track signals that switch between operating modes.

The second direction focuses on the mismatch that arises when the temporal resolution of the training data differs from that of the test data. We propose principled, zero-shot adaptation methods that recover the performance lost to this mismatch by adjusting only the neurons' dynamics parameters. We test the methods both in the single-device setting and in a multi-device federated learning framework aggregating models from devices operating at heterogeneous temporal resolutions.

Place, publisher, year, edition, pages
Uppsala: Acta Universitatis Upsaliensis, 2026. , p. 68
Series
Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology, ISSN 1651-6214 ; 2706
Keywords [en]
Spiking Neural Networks, State-Space Models, Signal Processing, Machine learning, Neuromoprhic
National Category
Engineering and Technology
Research subject
Electrical Engineering with specialization in Signal Processing
Identifiers
URN: urn:nbn:se:uu:diva-595106ISBN: 978-91-513-2921-5 (print)OAI: oai:DiVA.org:uu-595106DiVA, id: diva2:2090381
Public defence
2026-10-16, Häggsalen, Ångströmlaboratoriet, Lägerhyddsvägen 1, Uppsala, 09:00 (English)
Opponent
Supervisors
Available from: 2026-09-07 Created: 2026-08-06 Last updated: 2026-09-07
List of papers
1. State-Space Model Inspired Multiple-Input Multiple-Output Spiking Neurons
Open this publication in new window or tab >>State-Space Model Inspired Multiple-Input Multiple-Output Spiking Neurons
2025 (English)In: 2025 Neuro Inspired Computational Elements (NICE), Institute of Electrical and Electronics Engineers (IEEE), 2025Conference paper, Published paper (Refereed)
Abstract [en]

In spiking neural networks (SNNs), the main unit of information processing is the neuron with an internal state. The internal state generates an output spike based on its component associated with the membrane potential. This spike is then communicated to other neurons in the network. Here, we propose a general multiple-input multiple-output (MIMO) spiking neuron model that goes beyond this traditional single-input single-output (SISO) model in the SNN literature. Our proposed framework is based on interpreting the neurons as state-space models (SSMs) with linear state evolutions and non-linear spiking activation functions. We illustrate the trade-offs among various parameters of the proposed SSM-inspired neuron model, such as the number of hidden neuron states, the number of input and output channels, including single-input multiple-output (SIMO) and multiple-input single-output (MISO) models. We show that for SNNs with a small number of neurons with large internal state spaces, significant performance gains may be obtained by increasing the number of output channels of a neuron. In particular, a network with spiking neurons with multiple-output channels may achieve the same level of accuracy with the baseline with the continuous-valued communications on the same reference network architecture.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2025
Keywords
spiking neural networks (SNN), state-space models (SSM), multiple-input multiple-output (MIMO), neuromorphic
National Category
Control Engineering
Identifiers
urn:nbn:se:uu:diva-577312 (URN)10.1109/NICE65350.2025.11065909 (DOI)001553129800013 ()2-s2.0-105011414372 (Scopus ID)979-8-3315-0302-4 (ISBN)979-8-3315-0303-1 (ISBN)
Conference
2025 Neuro Inspired Computational Elements (NICE), 25-28 March, 2025, Heidelberg, Germany
Available from: 2026-01-22 Created: 2026-01-22 Last updated: 2026-08-06Bibliographically approved
2. Low-Bit Data Processing Using Multiple-Output Spiking Neurons With Non-Linear Reset Feedback
Open this publication in new window or tab >>Low-Bit Data Processing Using Multiple-Output Spiking Neurons With Non-Linear Reset Feedback
2025 (English)In: IEEE Journal on Selected Topics in Signal Processing, ISSN 1932-4553, E-ISSN 1941-0484, Vol. 19, no 6, p. 1172-1186Article in journal (Refereed) Published
Abstract [en]

Neuromorphic computing is an emerging technology enabling low-latency and energy-efficient signal processing. A key algorithmic tool in neuromorphic computing is spiking neural networks (SNNs). SNNs are biologically inspired neural networks which utilize stateful neurons, and provide low-bit data processing by encoding and decoding information using spikes. Similar to SNNs, deep state-space models (SSMs) utilize stateful building blocks. However, deep SSMs, which recently achieved competitive performance in various temporal modeling tasks, are typically designed with high-precision activation functions and no reset mechanisms. To bridge the gains offered by SNNs and the recent deep SSM models, we propose a novel multiple-output spiking neuron model that combines a linear, general SSM state transition with a non-linear feedback mechanism through reset. Compared to the existing neuron models for SNNs, our proposed model clearly conceptualizes the differences between the spiking function, the reset condition and the reset action. The experimental results on various tasks, i.e., a keyword spotting task, an event-based vision task and a sequential pattern recognition task, show that our proposed model achieves performance comparable to existing benchmarks in the SNN literature. Our results illustrate how the proposed reset mechanism can overcome instability and enable learning even when the linear part of neuron dynamics is unstable, allowing us to go beyond the strictly enforced stability of linear dynamics in recent deep SSM models.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2025
National Category
Engineering and Technology
Identifiers
urn:nbn:se:uu:diva-577307 (URN)10.1109/jstsp.2025.3595030 (DOI)001652373400009 ()2-s2.0-105012389845 (Scopus ID)
Funder
Swedish Research Council, 2022-06725
Available from: 2026-01-22 Created: 2026-01-22 Last updated: 2026-10-01Bibliographically approved
3. Delays in Spiking Neural Networks: A State Space Model Approach
Open this publication in new window or tab >>Delays in Spiking Neural Networks: A State Space Model Approach
2026 (English)Conference paper, Published paper (Refereed)
Abstract [en]

Spiking neural networks (SNNs) are biologically inspired, event-driven models suited for temporal data processing and energy-efficient neuromorphic computing. In SNNs, richer neuronal dynamic allows capturing more complex temporal dependencies, with delays playing a crucial role by allowing past inputs to directly influence present spiking behavior. We propose a general framework for incorporating delays into SNNs through additional state variables. The proposed mechanism enables each neuron to access a finite temporal input history. The framework is agnostic to neuron models and hence can be seamlessly integrated into standard spiking neuron models such as Leaky Integrate-and-Fire (LIF) and Adaptive LIF (adLIF). We analyze how the duration of the delays and the learnable parameters associated with them affect the performance. We investigate the trade-offs in the network architecture due to additional state variables introduced by the delay mechanism. Experiments on the Spiking Heidelberg Digits (SHD) dataset show that the proposed mechanism matches existing delay-based SNNs in performance while remaining computationally efficient, with particular gains in smaller networks.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2026
Keywords
spiking neural networks (SNN), state-space models (SSM), neuromorphic, delays, memory
National Category
Bioinformatics (Computational Biology)
Identifiers
urn:nbn:se:uu:diva-595083 (URN)
Conference
2026 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), 28 Sep-1 Oct, 2026, Atlanta, USA
Available from: 2026-08-06 Created: 2026-08-06 Last updated: 2026-08-11
4. Time-Varying Deep State Space Models forSequences with Switching Dynamics
Open this publication in new window or tab >>Time-Varying Deep State Space Models forSequences with Switching Dynamics
2026 (English)Conference paper, Published paper (Refereed)
Abstract [en]

The identification and modeling of time-varying systems is a fundamental challenge in signal processing and system identification. To address this challenge, we propose a class of time-varying state-space model (SSM) based neural networks in which the neurons' states are governed by time-varying dynamics. The proposed model provides the learnable time-varying dynamics through a dictionary of basis functions, where each basis function evolves differently over time. We evaluate the proposed approach on both synthetic data from switching systems and a speech denoising task where real audio is corrupted with switching dynamics noise. The results show that the proposed time-varying model consistently outperforms its time-invariant counterparts while maintaining comparable computational complexity. Our investigations also reveal which aspects of the time-varying dynamics of the data most need to be captured by the proposed time-invariant models, how the additional freedom provided by time-varying basis functions should be allocated across model components, and to what extent larger models can compensate for time-invariant limitations.

Keywords
System identification, Time-varying, Deep learning, Switching State-Space Models, Deep State-Space Models
National Category
Control Engineering
Identifiers
urn:nbn:se:uu:diva-595081 (URN)
Conference
23rd International Federation of Automatic Control (IFAC) World Congress, 23-28 August, 2026, Busan, Republic of Korea
Available from: 2026-08-06 Created: 2026-08-06 Last updated: 2026-08-11
5. Zero-shot Temporal Resolution Domain Adaptation for Spiking Neural Networks
Open this publication in new window or tab >>Zero-shot Temporal Resolution Domain Adaptation for Spiking Neural Networks
2026 (English)In: Neural Networks, ISSN 0893-6080, E-ISSN 1879-2782, Vol. 199, article id 108483Article in journal (Refereed) Published
Abstract [en]

Spiking Neural Networks (SNNs) are biologically-inspired deep neural networks that efficiently extract temporal information while offering promising gains in terms of energy efficiency and latency when deployed on neuromorphic devices. SNN parameters are sensitive to temporal resolution, leading to significant performance drops when the temporal resolution of target data during deployment is not the same as that of the source data used for training, especially when fine-tuning with the target data is not possible during deployment. To address this challenge, we propose three novel domain adaptation methods for adapting neuron parameters to account for the change in time resolution without re-training on target time resolution. The proposed methods are based on a mapping between neuron dynamics in SNNs and State Space Models (SSMs) and are applicable to general neuron models. We evaluate the proposed methods under spatio-temporal data tasks, namely the audio keyword spotting datasets SHD and MSWC, and the neuromorphic image NMINST dataset. Our methods provide an alternative to-and in most cases significantly outperform-the existing reference method that consists of scaling only the time constant. Notably, when the temporal resolution of the target data is double that of the source data, applying one of our proposed methods instead of the benchmark achieves classification accuracy of 89.5% instead of 53.0% on SHD, 93.6% instead of 38.8% on MSWC and 98.5% instead of 97.2% on NMNIST. Moreover, our results show that high accuracy on high temporal resolution data can be obtained by time-efficient training on lower temporal resolution data.

Place, publisher, year, edition, pages
Elsevier, 2026
Keywords
domain adaptation, temporal resolution, neuromorphic, spiking neural networks (SNNs), state-space models (SSMs)
National Category
Computer graphics and computer vision
Identifiers
urn:nbn:se:uu:diva-577313 (URN)10.1016/j.neunet.2025.108483 (DOI)001705451200001 ()
Funder
Swedish Research Council, 2024-05194Swedish Research Council, 2022-06725Uppsala University
Available from: 2026-01-22 Created: 2026-01-22 Last updated: 2026-08-06Bibliographically approved
6. Federated Learning of Spiking Neural Networks under Heterogeneous Temporal Resolutions
Open this publication in new window or tab >>Federated Learning of Spiking Neural Networks under Heterogeneous Temporal Resolutions
(English)Manuscript (preprint) (Other academic)
National Category
Computer Sciences Computer Engineering
Identifiers
urn:nbn:se:uu:diva-595104 (URN)
Available from: 2026-08-06 Created: 2026-08-06 Last updated: 2026-08-11Bibliographically approved
7. Spike timing–based coding in neuromimetic tactile system enables dynamic object classification
Open this publication in new window or tab >>Spike timing–based coding in neuromimetic tactile system enables dynamic object classification
Show others...
2024 (English)In: Science, ISSN 0036-8075, E-ISSN 1095-9203, Vol. 384, no 6696, p. 660-665Article in journal (Refereed) Published
Abstract [en]

Rapid processing of tactile information is essential to human haptic exploration and dexterous object manipulation. Conventional electronic skins generate frames of tactile signals upon interaction with objects. Unfortunately, they are generally ill-suited for efficient coding of temporal information and rapid feature extraction. In this work, we report a neuromorphic tactile system that uses spike timing, especially the first-spike timing, to code dynamic tactile information about touch and grasp. This strategy enables the system to seamlessly code highly dynamic information with millisecond temporal resolution on par with the biological nervous system, yielding dynamic extraction of tactile features. Upon interaction with objects, the system rapidly classifies them in the initial phase of touch and grasp, thus paving the way to fast tactile feedback desired for neuro-robotics and neuro-prosthetics. 

Place, publisher, year, edition, pages
American Association for the Advancement of Science (AAAS), 2024
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Artificial Intelligence; Electrical Engineering with Specialisation in Networked Embedded Systems; Electrical Engineering with specialization in Signal Processing; Engineering Science with specialization in Biomedical Engineering; Machine learning
Identifiers
urn:nbn:se:uu:diva-528024 (URN)10.1126/science.adf3708 (DOI)001253561400013 ()
Funder
EU, Horizon 2020, 965044Swedish Foundation for Strategic Research, FUS21-0067Swedish Research Council, 2019-05484
Available from: 2024-05-14 Created: 2024-05-14 Last updated: 2026-08-06Bibliographically approved

Open Access in DiVA

UUThesis_Karilanova,S-2026(1140 kB)220 downloads
File information
File name FULLTEXT01.pdfFile size 1140 kBChecksum SHA-512
181126c3ea2de8733f189bfefc90062c89869cca1e86eeca25e151b3c3ce2327ed4c2517f5a4894153f7f654ba662f7eff9c8913f3d43deb7645e1494ae31110
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Karilanova, Sanja
By organisation
Signals and Systems
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 220 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 1856 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf