Searcharxiv⌕ Search

arXiv subjects

Mihai A. Petrovici

Publications and source records attributed to Mihai A. Petrovici.

At least 19 recordsLinked to original sources
arXiv2026

ELiSe: Efficient Learning of Sequences in Structured Recurrent Networks

Behavior can be described as a temporal sequence of actions driven by neural activity. To learn complex sequential patterns in neural networks, memories of past activities need to persist on significantly longer timescales than the relaxation times of single-neuron activity. While recurrent networks can produce such long transients, training these networks is a challenge. Learning via error propagation confers models such as FORCE, RTRL or BPTT a significant functional advantage, but at the expense of biological plausibility. While reservoir computing circumvents this issue by learning only the readout weights, it does not scale well with problem complexity. We propose that two prominent structural features of cortical networks can alleviate these issues: the presence of a certain network scaffold at the onset of learning and the existence of dendritic compartments for enhancing neuronal information storage and computation. Our resulting model for Efficient Learning of Sequences (ELiSe) builds on these features to acquire and replay complex non-Markovian spatio-temporal patterns using only local, always-on and phase-free synaptic plasticity. We showcase the capabilities of ELiSe in a mock-up of birdsong learning, and demonstrate its flexibility with respect to parametrization, as well as its robustness to external disturbances.

q-bio.NC↗
arXiv2026

Neuromorphic architectures as numerical solvers for computational neuroscience

Neuromorphic computing is closely associated with spiking neuronal networks. However, an alternative class of so-called "rate-based" models arising from computational neuroscience and machine learning forgoes spiking interactions and instead relies on continuous coupling between neurons. Existing neuromorphic implementations designed around spike-based interactions are not well-suited for emulating such models. Here view the distributed simulation of these models as message-passing algorithms on parallel hardware. Leveraging prior art in numerical algorithms and distributed simulation, we outline steps that enable the design of efficient digital neuromorphic accelerators for non-spiking neuronal models. In particular, we show that multi-bit packets, rather than spikes, are the most efficient communication strategy in packet-switched networks and that compared to basic numerical integration methods, higher-order differential equation solvers decrease both computation and communication costs while achieving lower numerical error, but that these benefits are ultimately limited by arithmetic precision. Using our proposed design principles, we convert an existing neuromorphic architecture into a distributed numerical solver - a spikeless neuromorphic system - for continuously-coupled neuronal models. We thereby demonstrate that our theoretical considerations indeed translate into practical advantages, namely reduced energy consumption and delay.

cs.AR↗
arXiv2026

Dendritic structure enables powerful plasticity

Over the past decades, it has become increasingly clear that the complex morphology of cortical neurons is more than just a quirk of evolution, and that dendritic compartments serve as computational elements in their own right, rather than just providing connections between nerve cell bodies. While most computational studies discuss the enhanced representational capabilities of multi-compartment models as compared to point neurons, we focus here on the implications of neuronal morphology for synaptic plasticity. We argue that the ability of single neurons to simultaneously encode multiple pieces of information gives synapses local access to more than just the classical Hebbian pre- and postsynaptic terms, and with much greater specificity and reaction speed than permitted by other globally modulated factors. Based on a comparative review of recent dendritic learning models, we show how such neuronal compartmentalization can provide synapses with the means for calculating various forms of error signals, which in turn give rise to powerful real-time and fully local instantiations of deep learning through gradient descent. Implemented within cortical microcircuits capable of propagating and manipulating these errors, compartmentalized neurons thus ultimately enable the learning of far more complex tasks than are achievable by globally modulated Hebbian plasticity alone.

q-bio.NC↗
arXiv2026

Spike-based alignment learning solves the weight transport problem

In both machine learning and in computational neuroscience, plasticity in functional neural networks is frequently expressed as gradient descent on a cost. Often, this imposes symmetry constraints that are difficult to reconcile with local computation, as is required for biological networks or neuromorphic hardware. For example, wake-sleep learning in networks characterized by Boltzmann distributions assumes symmetric connectivity. Similarly, the error backpropagation algorithm is notoriously plagued by the weight transport problem between the representation and the error stream. Existing solutions such as feedback alignment circumvent the problem by deferring to the robustness of these algorithms to weight asymmetry. However, they scale poorly with network size and depth. We introduce spike-based alignment learning (SAL), a complementary learning rule for spiking neural networks, which uses spike timing statistics to extract and correct the asymmetry between effective reciprocal connections. Apart from being spike-based and fully local, our proposed mechanism takes advantage of noise. Based on an interplay between Hebbian and anti-Hebbian plasticity, synapses can thereby recover the true local gradient. This also alleviates discrepancies that arise from neuron and synapse variability -- an omnipresent property of physical neuronal networks. We demonstrate the efficacy of our mechanism using different spiking network models. First, SAL can significantly improve convergence to the target distribution in probabilistic spiking networks versus Hebbian plasticity alone. Second, in neuronal hierarchies based on cortical microcircuits, SAL effectively aligns feedback weights to the forward pathway, thus allowing the backpropagation of correct feedback errors. Third, our approach enables competitive performance in deep networks using only local plasticity for weight transport.

q-bio.NC↗
arXiv2026

Neural mechanisms of predictive processing: a collaborative community experiment through the OpenScope program

This review synthesizes advances in predictive processing within the sensory cortex. Predictive processing theorizes that the brain continuously predicts sensory inputs, refining neuronal responses by highlighting prediction errors. We identify key computational primitives, such as stimulus adaptation, dendritic computation, excitatory/inhibitory balance and hierarchical processing, as central to this framework. Our review highlights convergences, such as top-down inputs and inhibitory interneurons shaping mismatch signals, and divergences, including species-specific hierarchies and modality-dependent layer roles. To address these conflicts, we propose experiments in mice and primates using in-vivo two-photon imaging and electrophysiological recordings to test whether temporal, motor, and omission mismatch stimuli engage shared or distinct mechanisms. The resulting dataset, collected and shared via the OpenScope program, will enable model validation and community analysis, fostering iterative refinement and refutability to decode the neural circuits of predictive processing.

q-bio.NC↗
arXiv2026

A Variational Latent Equilibrium for Learning in Neuronal Circuits

Brains remain unrivaled in their ability to recognize and generate complex spatiotemporal patterns. While AI is able to reproduce some of these capabilities, deep learning algorithms remain largely at odds with our current understanding of brain circuitry and dynamics. This is prominently the case for backpropagation through time (BPTT), the go-to algorithm for learning complex temporal dependencies. In this work we propose a general formalism to approximate BPTT in a controlled, biologically plausible manner. Our approach builds on, unifies and extends several previous approaches to local, time-continuous, phase-free spatiotemporal credit assignment based on principles of energy conservation and extremal action. Our starting point is a prospective energy function of neuronal states, from which we calculate real-time error dynamics for time-continuous neuronal networks. In the general case, this provides a simple and straightforward derivation of the adjoint method result for neuronal networks, the time-continuous equivalent to BPTT. With a few modifications, we can turn this into a fully local (in space and time) set of equations for neuron and synapse dynamics. Our theory provides a rigorous framework for spatiotemporal deep learning in the brain, while simultaneously suggesting a blueprint for physical circuits capable of carrying out these computations. These results reframe and extend the recently proposed Generalized Latent Equilibrium (GLE) model.

q-bio.NC↗
arXiv2026

Real-time processing of analog signals on accelerated neuromorphic hardware

Sensory processing with neuromorphic systems is typically done by using either event-based sensors or translating input signals to spikes before presenting them to the neuromorphic processor. Here, we offer an alternative approach: direct analog signal injection eliminates superfluous and power-intensive analog-to-digital and digital-to-analog conversions, making it particularly suitable for efficient near-sensor processing. We demonstrate this by using the accelerated BrainScaleS-2 mixed-signal neuromorphic research platform and interfacing it directly to microphones and a servo-motor-driven actuator. Utilizing BrainScaleS-2's 1000-fold acceleration factor, we employ a spiking neural network to transform interaural time differences into a spatial code and thereby predict the location of sound sources. Our primary contributions are the first demonstrations of direct, continuous-valued sensor data injection into the analog compute units of the BrainScaleS-2 ASIC, and actuator control using its embedded microprocessors. This enables a fully on-chip processing pipeline$\unicode{x2014}$from sensory input handling, via spiking neural network processing to physical action. We showcase this by programming the system to localize and align a servo motor with the spatial direction of transient noise peaks in real-time.

cs.NE↗
arXiv2026

Backpropagation through space, time, and the brain

How physical networks of neurons, bound by spatio-temporal locality constraints, can perform efficient credit assignment, remains, to a large extent, an open question. In machine learning, the answer is almost universally given by the error backpropagation algorithm, through both space and time. However, this algorithm is well-known to rely on biologically implausible assumptions, in particular with respect to spatio-temporal (non-)locality. Alternative forward-propagation models such as real-time recurrent learning only partially solve the locality problem, but only at the cost of scaling, due to prohibitive storage requirements. We introduce Generalized Latent Equilibrium (GLE), a computational framework for fully local spatio-temporal credit assignment in physical, dynamical networks of neurons. We start by defining an energy based on neuron-local mismatches, from which we derive both neuronal dynamics via stationarity and parameter dynamics via gradient descent. The resulting dynamics can be interpreted as a real-time, biologically plausible approximation of backpropagation through space and time in deep cortical networks with continuous-time neuronal dynamics and continuously active, local synaptic plasticity. In particular, GLE exploits the morphology of dendritic trees to enable more complex information storage and processing in single neurons, as well as the ability of biological neurons to phase-shift their output rate with respect to their membrane potential, which is essential in both directions of information propagation. For the forward computation, it enables the mapping of time-continuous inputs to neuronal space, effectively performing a spatio-temporal convolution. For the backward computation, it permits the temporal inversion of feedback signals, which consequently approximate the adjoint variables necessary for useful parameter updates.

q-bio.NC↗
arXiv2025

Ultrafast neural sampling with spiking nanolasers

Owing to their significant advantages in terms of bandwidth, power efficiency, and latency, optical neuromorphic systems have arisen as interesting alternatives to digital electronic devices. Recently, photonic crystal nanolasers with excitable behavior were first demonstrated. Depending on the pumping strength, they emit short optical pulses -- spikes -- at various intervals on a nanosecond timescale. In this theoretical work, we show how networks of such photonic spiking neurons can be used for Bayesian inference through sampling from learned probability distributions. We provide a detailed derivation of translation rules from conventional sampling networks, such as Boltzmann machines, to photonic spiking networks and demonstrate their functionality across a range of generative tasks. Finally, we provide estimates of processing speed and power consumption, for which we expect improvements of several orders of magnitude over current state-of-the-art neuromorphic systems.

physics.optics↗
arXiv2025

Synchronization and semantization in deep spiking networks

Recent studies have shown how spiking networks can learn complex functionality through error-correcting plasticity, but the resulting structures and dynamics remain poorly studied. To elucidate how these models may link to observed dynamics in vivo and thus how they may ultimately explain cortical computation, we need a better understanding of their emerging patterns. We train a multi-layer spiking network, as a conceptual analog of the bottom-up visual hierarchy, for visual input classification using spike-time encoding. After learning, we observe the development of distinct spatio-temporal activity patterns. While input patterns are synchronous by construction, activity in early layers first spreads out over time, followed by re-convergence into sharp pulses as classes are gradually extracted. The emergence of synchronicity is accompanied by the formation of increasingly distinct pathways, reflecting the gradual semantization of input activity. We thus observe hierarchical networks learning spike latency codes to naturally acquire activity patterns characterized by synchronicity and separability, with pronounced excitatory pathways ascending through the layers. This provides a rigorous computational hypothesis for the experimentally observed synchronicity in the visual system as a natural consequence of deep learning in cortex.

q-bio.NC↗
arXiv2025

DelGrad: Exact event-based gradients for training delays and weights on spiking neuromorphic hardware

Spiking neural networks (SNNs) inherently rely on the timing of signals for representing and processing information. Incorporating trainable transmission delays, alongside synaptic weights, is crucial for shaping these temporal dynamics. While recent methods have shown the benefits of training delays and weights in terms of accuracy and memory efficiency, they rely on discrete time, approximate gradients, and full access to internal variables like membrane potentials. This limits their precision, efficiency, and suitability for neuromorphic hardware due to increased memory requirements and I/O bandwidth demands. To address these challenges, we propose DelGrad, an analytical, event-based method to compute exact loss gradients for both synaptic weights and delays. The inclusion of delays in the training process emerges naturally within our proposed formalism, enriching the model's search space with a temporal dimension. Moreover, DelGrad, grounded purely in spike timing, eliminates the need to track additional variables such as membrane potentials. To showcase this key advantage, we demonstrate the functionality and benefits of DelGrad on the BrainScaleS-2 neuromorphic platform, by training SNNs in a chip-in-the-loop fashion. For the first time, we experimentally demonstrate the memory efficiency and accuracy benefits of adding delays to SNNs on noisy mixed-signal hardware. Additionally, these experiments also reveal the potential of delays for stabilizing networks against noise. DelGrad opens a new way for training SNNs with delays on neuromorphic hardware, which results in fewer required parameters, higher accuracy and ease of hardware training.

cs.NE↗
arXiv2025

NeuroBench: A Framework for Benchmarking Neuromorphic Computing Algorithms and Systems

Neuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks standardized benchmarks, making it difficult to accurately measure technological advancements, compare performance with conventional methods, and identify promising future research directions. Prior neuromorphic computing benchmark efforts have not seen widespread adoption due to a lack of inclusive, actionable, and iterative benchmark design and guidelines. To address these shortcomings, we present NeuroBench: a benchmark framework for neuromorphic computing algorithms and systems. NeuroBench is a collaboratively-designed effort from an open community of researchers across industry and academia, aiming to provide a representative structure for standardizing the evaluation of neuromorphic approaches. The NeuroBench framework introduces a common set of tools and systematic methodology for inclusive benchmark measurement, delivering an objective reference framework for quantifying neuromorphic approaches in both hardware-independent (algorithm track) and hardware-dependent (system track) settings. In this article, we outline tasks and guidelines for benchmarks across multiple application domains, and present initial performance baselines across neuromorphic and conventional approaches for both benchmark tracks. NeuroBench is intended to continually expand its benchmarks and features to foster and track the progress made by the research community.

cs.AI↗
arXiv2024

Lu.i -- A low-cost electronic neuron for education and outreach

With an increasing presence of science throughout all parts of society, there is a rising expectation for researchers to effectively communicate their work and, equally, for teachers to discuss contemporary findings in their classrooms. While the community can resort to an established set of teaching aids for the fundamental concepts of most natural sciences, there is a need for similarly illustrative experiments and demonstrators in neuroscience. We therefore introduce Lu.i: a parametrizable electronic implementation of the leaky-integrate-and-fire neuron model in an engaging form factor. These palm-sized neurons can be used to visualize and experience the dynamics of individual cells and small spiking neural networks. When stimulated with real or simulated sensory input, Lu.i demonstrates brain-inspired information processing in the hands of a student. As such, it is actively used at workshops, in classrooms, and for science communication. As a versatile tool for teaching and outreach, Lu.i nurtures the comprehension of neuroscience research and neuromorphic engineering among future generations of scientists and in the general public.

q-bio.NC↗
arXiv2024

Confidence and second-order errors in cortical circuits

Minimization of cortical prediction errors has been considered a key computational goal of the cerebral cortex underlying perception, action and learning. However, it is still unclear how the cortex should form and use information about uncertainty in this process. Here, we formally derive neural dynamics that minimize prediction errors under the assumption that cortical areas must not only predict the activity in other areas and sensory streams but also jointly project their confidence (inverse expected uncertainty) in their predictions. In the resulting neuronal dynamics, the integration of bottom-up and top-down cortical streams is dynamically modulated based on confidence in accordance with the Bayesian principle. Moreover, the theory predicts the existence of cortical second-order errors, comparing confidence and actual performance. These errors are propagated through the cortical hierarchy alongside classical prediction errors and are used to learn the weights of synapses responsible for formulating confidence. We propose a detailed mapping of the theory to cortical circuitry, discuss entailed functional interpretations and provide potential directions for experimental work.

q-bio.NC↗
arXiv2024

Learning efficient backprojections across cortical hierarchies in real time

Models of sensory processing and learning in the cortex need to efficiently assign credit to synapses in all areas. In deep learning, a known solution is error backpropagation, which however requires biologically implausible weight transport from feed-forward to feedback paths. We introduce Phaseless Alignment Learning (PAL), a bio-plausible method to learn efficient feedback weights in layered cortical hierarchies. This is achieved by exploiting the noise naturally found in biophysical systems as an additional carrier of information. In our dynamical system, all weights are learned simultaneously with always-on plasticity and using only information locally available to the synapses. Our method is completely phase-free (no forward and backward passes or phased learning) and allows for efficient error propagation across multi-layer cortical hierarchies, while maintaining biologically plausible signal transport and learning. Our method is applicable to a wide class of models and improves on previously known biologically plausible ways of credit assignment: compared to random synaptic feedback, it can solve complex tasks with less neurons and learn more useful latent representations. We demonstrate this on various classification tasks using a cortical microcircuit model with prospective coding.

q-bio.NC↗
arXiv2023

Emulating insect brains for neuromorphic navigation

Bees display the remarkable ability to return home in a straight line after meandering excursions to their environment. Neurobiological imaging studies have revealed that this capability emerges from a path integration mechanism implemented within the insect's brain. In the present work, we emulate this neural network on the neuromorphic mixed-signal processor BrainScaleS-2 to guide bees, virtually embodied on a digital co-processor, back to their home location after randomly exploring their environment. To realize the underlying neural integrators, we introduce single-neuron spike-based short-term memory cells with axo-axonic synapses. All entities, including environment, sensory organs, brain, actuators, and the virtual body, run autonomously on a single BrainScaleS-2 microchip. The functioning network is fine-tuned for better precision and reliability through an evolution strategy. As BrainScaleS-2 emulates neural processes 1000 times faster than biology, 4800 consecutive bee journeys distributed over 320 generations occur within only half an hour on a single neuromorphic core.

cs.NE↗
arXiv2023

Learning beyond sensations: how dreams organize neuronal representations

Semantic representations in higher sensory cortices form the basis for robust, yet flexible behavior. These representations are acquired over the course of development in an unsupervised fashion and continuously maintained over an organism's lifespan. Predictive learning theories propose that these representations emerge from predicting or reconstructing sensory inputs. However, brains are known to generate virtual experiences, such as during imagination and dreaming, that go beyond previously experienced inputs. Here, we suggest that virtual experiences may be just as relevant as actual sensory inputs in shaping cortical representations. In particular, we discuss two complementary learning principles that organize representations through the generation of virtual experiences. First, "adversarial dreaming" proposes that creative dreams support a cortical implementation of adversarial learning in which feedback and feedforward pathways engage in a productive game of trying to fool each other. Second, "contrastive dreaming" proposes that the invariance of neuronal representations to irrelevant factors of variation is acquired by trying to map similar virtual experiences together via a contrastive learning process. These principles are compatible with known cortical structure and dynamics and the phenomenology of sleep thus providing promising directions to explain cortical learning beyond the classical predictive learning paradigm.

q-bio.NC↗
arXiv2023

Conductance-based dendrites perform Bayes-optimal cue integration

A fundamental function of cortical circuits is the integration of information from different sources to form a reliable basis for behavior. While animals behave as if they optimally integrate information according to Bayesian probability theory, the implementation of the required computations in the biological substrate remains unclear. We propose a novel, Bayesian view on the dynamics of conductance-based neurons and synapses which suggests that they are naturally equipped to optimally perform information integration. In our approach apical dendrites represent prior expectations over somatic potentials, while basal dendrites represent likelihoods of somatic potentials. These are parametrized by local quantities, the effective reversal potentials and membrane conductances. We formally demonstrate that under these assumptions the somatic compartment naturally computes the corresponding posterior. We derive a gradient-based plasticity rule, allowing neurons to learn desired target distributions and weight synaptic inputs by their relative reliabilities. Our theory explains various experimental findings on the system and single-cell level related to multi-sensory integration, which we illustrate with simulations. Furthermore, we make experimentally testable predictions on Bayesian dendritic integration and synaptic plasticity.

q-bio.NC↗