Johannes Leugering
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Johannes Leugering has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Gordon Pipa
Institute of Cognitive Science, Osnabrueck University, 49074 Osnabrueck, Germany
4 shared publications - Gert Cauwenberghs
University of California, San Diego
3 shared publications - Shantanu Chakrabartty
Department of Electrical and Systems Engineering, Washington University in St. Louis, St. Louis, USA
3 shared publications - Zihao Chen
Department of Electrical and Systems Engineering, Washington University in St. Louis, St. Louis, USA
3 shared publications Adil Malik
Department of Electrical and Electronic Engineering, Imperial College London, London, UK
2 shared publicationsAndreas G. Andreou
Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, USA
2 shared publications
Latest Publications
- 30 March 2026
- Published by American Physical Society (APS) in Physical Review E
- Vol. 113 (3) , 035311
Estimation of energy-dissipation lower bounds for neuromorphic learning in memory
Neuromorphic or neurally inspired optimizers rely on local but parallel parameter updates to solve problems that range from quadratic programming to Ising machines. An ideal realization of such an optimizer not only uses a compute-in-memory paradigm to address the memory wall (i.e., energy dissipated due to repeated memory read access), but also uses a learning-in-memory paradigm to address the energy bottlenecks due to repeated memory writes at the precision required for optimization (the update wall), and to address the energy bottleneck due to the repeated transfer of information between short-term and long-term memories (the consolidation wall). In this paper, we derive theoretical estimates for the energy-to-solution metric that can be achieved by this ideal neuromorphic optimizer which is realized by modulating the energy barrier of the physical memories such that the dynamics of memory updates and memory consolidation matches the optimization or the annealing dynamics. The analysis presented in this paper captures the out-of-equilibrium thermodynamics of learning and the resulting energy-efficiency estimates are model agnostic and only depend on the number of model-update operations, the model size in terms of number of parameters, the speed of convergence, and the precision of the solution. To show the practical applicability of our results, we apply our analysis for estimating the lower bound on the energy-to-solution metrics for large-scale AI workloads.
- 30 March 2026
- Published by American Physical Society (APS) in Physical Review E
- Vol. 113 (3) , 035311
Estimation of energy-dissipation lower bounds for neuromorphic learning in memory
Neuromorphic or neurally inspired optimizers rely on local but parallel parameter updates to solve problems that range from quadratic programming to Ising machines. An ideal realization of such an optimizer not only uses a compute-in-memory paradigm to address the memory wall (i.e., energy dissipated due to repeated memory read access), but also uses a learning-in-memory paradigm to address the energy bottlenecks due to repeated memory writes at the precision required for optimization (the update wall), and to address the energy bottleneck due to the repeated transfer of information between short-term and long-term memories (the consolidation wall). In this paper, we derive theoretical estimates for the energy-to-solution metric that can be achieved by this ideal neuromorphic optimizer which is realized by modulating the energy barrier of the physical memories such that the dynamics of memory updates and memory consolidation matches the optimization or the annealing dynamics. The analysis presented in this paper captures the out-of-equilibrium thermodynamics of learning and the resulting energy-efficiency estimates are model agnostic and only depend on the number of model-update operations, the model size in terms of number of parameters, the speed of convergence, and the precision of the solution. To show the practical applicability of our results, we apply our analysis for estimating the lower bound on the energy-to-solution metrics for large-scale AI workloads.
- Open Access
HiAER-Spike Software-Hardware Reconfigurable Platform for Event-Driven Neuromorphic Computing at Scale
In this work, we present HiAER-Spike, a modular, reconfigurable, event-driven neuromorphic computing platform designed to execute large spiking neural networks with up to 160 million neurons and 40 billion synapses - roughly twice the neurons of a mouse brain at faster than real time. This system, assembled at the UC San Diego Supercomputer Center, comprises a co-designed hard- and software stack that is optimized for run-time massively parallel processing and hierarchical address-event routing (HiAER) of spikes while promoting memory-efficient network storage and execution. The architecture efficiently handles both sparse connectivity and sparse activity for robust and low-latency event-driven inference for both edge and cloud computing. A Python programming interface to HiAER-Spike, agnostic to hardware-level detail, shields the user from complexity in the configuration and execution of general spiking neural networks with minimal constraints in topology. The system is made easily available over a web portal for use by the wider community. In the following, we provide an overview of the hard- and software stack, explain the underlying design principles, demonstrate some of the system's capabilities and solicit feedback from the broader neuromorphic community. Examples are shown demonstrating HiAER-Spike's capabilities for event-driven vision on benchmark CIFAR-10, DVS event-based gesture, MNIST, and Pong tasks.
Open AccessHiAER-Spike Software-Hardware Reconfigurable Platform for Event-Driven Neuromorphic Computing at Scale
In this work, we present HiAER-Spike, a modular, reconfigurable, event-driven neuromorphic computing platform designed to execute large spiking neural networks with up to 160 million neurons and 40 billion synapses - roughly twice the neurons of a mouse brain at faster than real time. This system, assembled at the UC San Diego Supercomputer Center, comprises a co-designed hard- and software stack that is optimized for run-time massively parallel processing and hierarchical address-event routing (HiAER) of spikes while promoting memory-efficient network storage and execution. The architecture efficiently handles both sparse connectivity and sparse activity for robust and low-latency event-driven inference for both edge and cloud computing. A Python programming interface to HiAER-Spike, agnostic to hardware-level detail, shields the user from complexity in the configuration and execution of general spiking neural networks with minimal constraints in topology. The system is made easily available over a web portal for use by the wider community. In the following, we provide an overview of the hard- and software stack, explain the underlying design principles, demonstrate some of the system's capabilities and solicit feedback from the broader neuromorphic community. Examples are shown demonstrating HiAER-Spike's capabilities for event-driven vision on benchmark CIFAR-10, DVS event-based gesture, MNIST, and Pong tasks.
- Open Access
- 1 July 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1
Author Correction: ON-OFF neuromorphic ISING machines using Fowler-Nordheim annealers
- 1 July 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1
Open AccessAuthor Correction: ON-OFF neuromorphic ISING machines using Fowler-Nordheim annealers
- Open Access
- 31 March 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-13
ON-OFF neuromorphic ISING machines using Fowler-Nordheim annealers
We introduce NeuroSA, a neuromorphic architecture specifically designed to ensure asymptotic convergence to the ground state of an Ising problem using a Fowler-Nordheim quantum mechanical tunneling based threshold-annealing process. The core component of NeuroSA consists of a pair of asynchronous ON-OFF neurons, which effectively map classical simulated annealing dynamics onto a network of integrate-and-fire neurons. The threshold of each ON-OFF neuron pair is adaptively adjusted by an FN annealer and the resulting spiking dynamics replicates the optimal escape mechanism and convergence of SA, particularly at low-temperatures. To validate the effectiveness of our neuromorphic Ising machine, we systematically solved benchmark combinatorial optimization problems such as MAX-CUT and Max Independent Set. Across multiple runs, NeuroSA consistently generates distribution of solutions that are concentrated around the state-of-the-art results (within 99%) or surpass the current state-of-the-art solutions for Max Independent Set benchmarks. Furthermore, NeuroSA is able to achieve these superior distributions without any graph-specific hyperparameter tuning. For practical illustration, we present results from an implementation of NeuroSA on the SpiNNaker2 platform, highlighting the feasibility of mapping our proposed architecture onto a standard neuromorphic accelerator platform.
- 31 March 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-13
Open AccessON-OFF neuromorphic ISING machines using Fowler-Nordheim annealers
We introduce NeuroSA, a neuromorphic architecture specifically designed to ensure asymptotic convergence to the ground state of an Ising problem using a Fowler-Nordheim quantum mechanical tunneling based threshold-annealing process. The core component of NeuroSA consists of a pair of asynchronous ON-OFF neurons, which effectively map classical simulated annealing dynamics onto a network of integrate-and-fire neurons. The threshold of each ON-OFF neuron pair is adaptively adjusted by an FN annealer and the resulting spiking dynamics replicates the optimal escape mechanism and convergence of SA, particularly at low-temperatures. To validate the effectiveness of our neuromorphic Ising machine, we systematically solved benchmark combinatorial optimization problems such as MAX-CUT and Max Independent Set. Across multiple runs, NeuroSA consistently generates distribution of solutions that are concentrated around the state-of-the-art results (within 99%) or surpass the current state-of-the-art solutions for Max Independent Set benchmarks. Furthermore, NeuroSA is able to achieve these superior distributions without any graph-specific hyperparameter tuning. For practical illustration, we present results from an implementation of NeuroSA on the SpiNNaker2 platform, highlighting the feasibility of mapping our proposed architecture onto a standard neuromorphic accelerator platform.
- Open Access
- 20 November 2024
- Published by Wiley in Annals of the New York Academy of Sciences
- Vol. 1542 (1) , 5-10
Neuromorphic engineering: Artificial brains for artificial intelligence
Neuromorphic engineering is a research discipline that tries to bridge the gaps between neuroscience and engineering, cognition and algorithms, and natural and artificial intelligence. Neuromorphic engineering promises revolutionary breakthroughs that could rapidly advance our understanding of the brain and pave the way toward more human‐like and sustainable artificial intelligence. But first, it will have to find its way out of the laboratory.
- 20 November 2024
- Published by Wiley in Annals of the New York Academy of Sciences
- Vol. 1542 (1) , 5-10
Open AccessNeuromorphic engineering: Artificial brains for artificial intelligence
Neuromorphic engineering is a research discipline that tries to bridge the gaps between neuroscience and engineering, cognition and algorithms, and natural and artificial intelligence. Neuromorphic engineering promises revolutionary breakthroughs that could rapidly advance our understanding of the brain and pave the way toward more human‐like and sustainable artificial intelligence. But first, it will have to find its way out of the laboratory.
- Open Access
- 18 December 2019
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 14 (12) , e0225838
A Bayesian Monte Carlo approach for predicting the spread of infectious diseases
In this paper, a simple yet interpretable, probabilistic model is proposed for the prediction of reported case counts of infectious diseases. A spatio-temporal kernel is derived from training data to capture the typical interaction effects of reported infections across time and space, which provides insight into the dynamics of the spread of infectious diseases. Testing the model on a one-week-ahead prediction task for campylobacteriosis and rotavirus infections across Germany, as well as Lyme borreliosis across the federal state of Bavaria, shows that the proposed model performs on-par with the state-of-the-art hhh4 model. However, it provides a full posterior distribution over parameters in addition to model predictions, which aides in the assessment of the model. The employed Bayesian Monte Carlo regression framework is easily extensible and allows for incorporating prior domain knowledge, which makes it suitable for use on limited, yet complex datasets as often encountered in epidemiology.
- 18 December 2019
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 14 (12) , e0225838
Open AccessA Bayesian Monte Carlo approach for predicting the spread of infectious diseases
In this paper, a simple yet interpretable, probabilistic model is proposed for the prediction of reported case counts of infectious diseases. A spatio-temporal kernel is derived from training data to capture the typical interaction effects of reported infections across time and space, which provides insight into the dynamics of the spread of infectious diseases. Testing the model on a one-week-ahead prediction task for campylobacteriosis and rotavirus infections across Germany, as well as Lyme borreliosis across the federal state of Bavaria, shows that the proposed model performs on-par with the state-of-the-art hhh4 model. However, it provides a full posterior distribution over parameters in addition to model predictions, which aides in the assessment of the model. The employed Bayesian Monte Carlo regression framework is easily extensible and allows for incorporating prior domain knowledge, which makes it suitable for use on limited, yet complex datasets as often encountered in epidemiology.
- Open Access
Dendritic plateau potentials can process spike sequences across multiple time-scales
A bstract: The brain constantly processes information encoded in temporal sequences of spiking activity. This sequential activity emerges from sensory inputs as well as from the brain’s own recurrent connectivity and spans multiple dynamically changing timescales. Decoding the temporal order of spiking activity across these varying timescales is a critical function of the brain, but we do not yet understand its neural implementation. The problem is, that the passive dynamics of neural membrane potentials occur on a short millisecond timescale, whereas many cognitive tasks require the integration of information across much slower behavioral timescales. However, actively generated dendritic plateau potentials do occur on such longer timescales, and their essential role for many aspects of cognition has been firmly established by recent experiments. Here, we build on these discoveries and propose a new model of neural computation that emerges from the interaction of localized plateau potentials across a functionally compartmentalized dendritic tree. We show how this interaction offers a robust solution to the timing invariant detection and processing of sequential spike patterns in single neurons. Stochastic synaptic transmission complements the deterministic all-or-none plateau process and improves information transmission by allowing ensembles of neurons to produce graded responses to continuous combinations of features. We found that networks of such neurons can solve highly complex sequence detection tasks by breaking down long inputs into sequences of shorter, random features that can be classified reliably. These results suggest that active dendritic processes are fundamental to neural computation.
Open AccessDendritic plateau potentials can process spike sequences across multiple time-scales
A bstract: The brain constantly processes information encoded in temporal sequences of spiking activity. This sequential activity emerges from sensory inputs as well as from the brain’s own recurrent connectivity and spans multiple dynamically changing timescales. Decoding the temporal order of spiking activity across these varying timescales is a critical function of the brain, but we do not yet understand its neural implementation. The problem is, that the passive dynamics of neural membrane potentials occur on a short millisecond timescale, whereas many cognitive tasks require the integration of information across much slower behavioral timescales. However, actively generated dendritic plateau potentials do occur on such longer timescales, and their essential role for many aspects of cognition has been firmly established by recent experiments. Here, we build on these discoveries and propose a new model of neural computation that emerges from the interaction of localized plateau potentials across a functionally compartmentalized dendritic tree. We show how this interaction offers a robust solution to the timing invariant detection and processing of sequential spike patterns in single neurons. Stochastic synaptic transmission complements the deterministic all-or-none plateau process and improves information transmission by allowing ensembles of neurons to produce graded responses to continuous combinations of features. We found that networks of such neurons can solve highly complex sequence detection tasks by breaking down long inputs into sequences of shorter, random features that can be classified reliably. These results suggest that active dendritic processes are fundamental to neural computation.
- 1 April 2018
- Published by MIT Press in Neural Computation
- Vol. 30 (4) , 945-986
A Unifying Framework of Synaptic and Intrinsic Plasticity in Neural Populations
A neuronal population is a computational unit that receives a multivariate, time-varying input signal and creates a related multivariate output. These neural signals are modeled as stochastic processes that transmit information in real time, subject to stochastic noise. In a stationary environment, where the input signals can be characterized by constant statistical properties, the systematic relationship between its input and output processes determines the computation carried out by a population. When these statistical characteristics unexpectedly change, the population needs to adapt to its new environment if it is to maintain stable operation. Based on the general concept of homeostatic plasticity, we propose a simple compositional model of adaptive networks that achieve invariance with regard to undesired changes in the statistical properties of their input signals and maintain outputs with well-defined joint statistics. To achieve such invariance, the network model combines two functionally distinct types of plasticity. An abstract stochastic process neuron model implements a generalized form of intrinsic plasticity that adapts marginal statistics, relying only on mechanisms locally confined within each neuron and operating continuously in time, while a simple form of Hebbian synaptic plasticity operates on synaptic connections, thus shaping the interrelation between neurons as captured by a copula function. The combined effect of both mechanisms allows a neuron population to discover invariant representations of its inputs that remain stable under a wide range of transformations (e.g., shifting, scaling and (affine linear) mixing). The probabilistic model of homeostatic adaptation on a population level as presented here allows us to isolate and study the individual and the interaction dynamics of both mechanisms of plasticity and could guide the future search for computationally beneficial types of adaptation.
- 1 April 2018
- Published by MIT Press in Neural Computation
- Vol. 30 (4) , 945-986
A Unifying Framework of Synaptic and Intrinsic Plasticity in Neural Populations
A neuronal population is a computational unit that receives a multivariate, time-varying input signal and creates a related multivariate output. These neural signals are modeled as stochastic processes that transmit information in real time, subject to stochastic noise. In a stationary environment, where the input signals can be characterized by constant statistical properties, the systematic relationship between its input and output processes determines the computation carried out by a population. When these statistical characteristics unexpectedly change, the population needs to adapt to its new environment if it is to maintain stable operation. Based on the general concept of homeostatic plasticity, we propose a simple compositional model of adaptive networks that achieve invariance with regard to undesired changes in the statistical properties of their input signals and maintain outputs with well-defined joint statistics. To achieve such invariance, the network model combines two functionally distinct types of plasticity. An abstract stochastic process neuron model implements a generalized form of intrinsic plasticity that adapts marginal statistics, relying only on mechanisms locally confined within each neuron and operating continuously in time, while a simple form of Hebbian synaptic plasticity operates on synaptic connections, thus shaping the interrelation between neurons as captured by a copula function. The combined effect of both mechanisms allows a neuron population to discover invariant representations of its inputs that remain stable under a wide range of transformations (e.g., shifting, scaling and (affine linear) mixing). The probabilistic model of homeostatic adaptation on a population level as presented here allows us to isolate and study the individual and the interaction dynamics of both mechanisms of plasticity and could guide the future search for computationally beneficial types of adaptation.
- 1 March 2017
- Published by IBM in IBM Journal of Research and Development
- Vol. 61 (2/3) , 8:7-8:9
Neuromorphic computation in multi-delay coupled models
- 1 March 2017
- Published by IBM in IBM Journal of Research and Development
- Vol. 61 (2/3) , 8:7-8:9
Neuromorphic computation in multi-delay coupled models