Benoit Cottereau

Three fully funded PhD fellowships are available to work at the CNRS (Cerco laboratory in Toulouse and also IPAL in Singapore) under the supervision of Benoit Cottereau (CNRS research director), in close collaboration with Robby Tan (assistant professor at NUS, Singapore). The aim of the project is to develop computational models of spatial cognition in humans (e.g., for depth/motion perception or navigation). In order to reproduce the very low energy consumption of visual information processing in biological systems, these models will be based on spiking-neural networks (SNNs) and will process data collected from event-based cameras. They should also adapt to variations in visual conditions (e.g., to night scenes or when parts of the input data are occluded). These models will be evaluated with respect to state-of-art approaches in computer vision/AI but also with respect to biological data (their responses will be compared to those observed in neural recordings and/or in behavioural experiments).

Benoit Cottereau

Fully funded PhD fellowships are available to work at the CNRS (Cerco laboratory in Toulouse and also IPAL in Singapore) under the supervision of Benoit Cottereau (CNRS research director), in close collaboration with Robby Tan (assistant professor at NUS, Singapore). The aim of the project is to develop computational models of spatial cognition in humans (e.g., for depth/motion perception or navigation). In order to reproduce the very low energy consumption of visual information processing in biological systems, these models will be based on spiking-neural networks (SNNs) and will process data collected from event-based cameras. These models will be evaluated with respect to biological data (their responses will be compared to those observed in neural recordings and/or in behavioural experiments).

Friedemann Zenke

The position involves conducting research in computational neuroscience and bio-inspired machine intelligence, writing research articles and presenting them at international conferences, publishing in neuroscience journals and machine learning venues such as ICML, NeurIPS, ICLR, etc., and interacting and collaborating with experimental neuroscience groups or neuromorphic hardware developers nationally and internationally.

N/A

Two Postdoctoral Research Associates in Neurorobotics are required for a period of 48 months to work on the Horizon/InnovateUK project “PRIMI: Performance in Robots Interaction via Mental Imagery. This is a collaborative project of the University of Manchester’s Cognitive Robotics Lab with various academic and industry partners in the UK and Europe. PRIMI will synergistically combine research and development in neurophysiology, psychology, machine intelligence, cognitive mechatronics, neuromorphic engineering, and humanoid robotics to build developmental models of higher-cognition abilities – mental imagery, abstract reasoning, and theory of mind – boosted by energy- efficient event-driven computing and sensing. You will carry out research on robot neuro/cognitive architectures, using a combination of machine learning and robotics methodologies. You will be working collaboratively as part of the Cognitive Robotics Lab at the Department of Computer Science at the University of Manchester under the supervision of Professor Angelo Cangelosi.

Ján Antolík

A postdoctoral position within the Computational Systems Neuroscience Group (CSNG) at Charles University, Prague, focusing on computational neuroscience and neuro-prosthetic system design. The group explores the intricacies of the visual system, sensory coding, and neuro-prosthetic solutions using computational approaches such as large-scale biologically detailed spiking network models, firing-rate models of development, and modern machine learning techniques. The team is dedicated to understanding visual perception and its restoration via neuro-prosthetic devices. Multiple project topics are available and can be adjusted to the interest and background of the applicant, including modeling electrical stimulation in a spiking model of the primary visual cortex, deep-neural networks in visual neuroscience, study of cortical dynamics in the visual cortex, and biologically detailed spiking large-scale models of early visual cortical pathway from Retina to V4.

El-ghazali TALBI

N/A

Benoit Cottereau / Timothée Masquelier

In primates, neuronal processing of visual information occurs within two main pathways. The ventral pathway (or ‘what’ pathway) underlies cognitive functions such as faces or object recognition whereas the dorsal pathway (or ‘where’ pathway) allows visually guided actions such as navigation or manipulation of objects. In recent years, numerous studies in computational neurosciences used deep neural networks to model neuronal processing within the ventral pathway. Much less is known about treatments carried out along the dorsal pathway. This multidisciplinary project at the interface between AI, computational neurosciences and biological vision aims at modelling motion and depth processing along the dorsal pathway of the primate visual system. In order to reproduce the very low energy consumption of the primate brain, the developed models will be based on spiking-neural networks (SNNs) and will process data collected from event-based cameras. We will be particularly interested in the processing of higher-level forms of motion (optic flow, motion-in-depth, biological motion). The responses of the neural networks after (supervised or unsupervised) training will be compared to biological data from the literature (for example with electrophysiological data recorded in non-human primates, see for example Duffy and Wurtz, 1991) or acquired in the CerCo laboratory.

Jie Mei

The Wiring, Neuromodeling and Brain Lab at IT:U Interdisciplinary Transformation University Austria is offering 2 PhD positions in neuromodulation-aware artificial intelligence. We are interested in (1) the role of individual neuromodulators (e.g., dopamine, serotonin, and acetylcholine) in initiating and implementing diverse biological and cognitive functions, (2) how competition and cooperation among neuromodulators enrich single neuromodulator computations, and (3) how multi-neuromodulator dynamics can be translated into learning rules for more flexible, robust, and adaptive learning in artificial neural networks.

Dr. Fleur Zeldenrust

For the Vidi project ‘Top-down neuromodulation and bottom-up network computation,’ we seek a postdoc to study neuromodulators in efficient spike-coding networks. Using our lab’s data on dopamine, acetylcholine, and serotonin from the mouse barrel cortex, you’ll derive models connecting single cells, networks, and behavior. The aim of this project is to explain the effects of neuromodulation on task performance in biologically realistic spiking recurrent neural networks (SRNNs). You will use the efficient spike coding framework, in which a network is not trained by a learning paradigm but deduced using mathematically rigorous rules that enforce efficient coding (i.e. maximally informative spikes). You will study how the network’s structural properties such as neural heterogeneity influence decoding performance and efficiency. You will incorporate realistic network properties of the (barrel) cortex based on our lab’s measurements and incorporate the cellular effects of dopamine, acetylcholine and serotonin we have measured over the past years into the network, to investigate their effects on representations, network activity measures such as dimensionality, and decoding performance. You will build on the single cell data, network models and analysis methods available in our group, and your results will be incorporated into our group’s further research to develop and validate efficient coding models of (somatosensory) perception. Therefore, we are looking for a team player who is willing to learn from the other group members and to share their knowledge with them.

Loss shaping enhances exact gradient learning with EventProp in Spiking Neural Networks

Bridging the gap between artificial models and cortical circuits

Artificial neural networks simplify complex biological circuits into tractable models for computational exploration and experimentation. However, the simplification of artificial models also undermines their applicability to real brain dynamics. Typical efforts to address this mismatch add complexity to increasingly unwieldy models. Here, we take a different approach; by reducing the complexity of a biological cortical culture, we aim to distil the essential factors of neuronal dynamics and plasticity. We leverage recent advances in growing neurons from human induced pluripotent stem cells (hiPSCs) to analyse ex vivo cortical cultures with only two distinct excitatory and inhibitory neuron populations. Over 6 weeks of development, we record from thousands of neurons using high-density microelectrode arrays (HD-MEAs) that allow access to individual neurons and the broader population dynamics. We compare these dynamics to two-population artificial networks of single-compartment neurons with random sparse connections and show that they produce similar dynamics. Specifically, our model captures the firing and bursting statistics of the cultures. Moreover, tightly integrating models and cultures allows us to evaluate the impact of changing architectures over weeks of development, with and without external stimuli. Broadly, the use of simplified cortical cultures enables us to use the repertoire of theoretical neuroscience techniques established over the past decades on artificial network models. Our approach of deriving neural networks from human cells also allows us, for the first time, to directly compare neural dynamics of disease and control. We found that cultures e.g. from epilepsy patients tended to have increasingly more avalanches of synchronous activity over weeks of development, in contrast to the control cultures. Next, we will test possible interventions, in silico and in vitro, in a drive for personalised approaches to medical care. This work starts bridging an important theoretical-experimental neuroscience gap for advancing our understanding of mammalian neuron dynamics.

A biologically plausible inhibitory plasticity rule for world-model learning in SNNs

Memory consolidation is the process by which recent experiences are assimilated into long-term memory. In animals, this process requires the offline replay of sequences observed during online exploration in the hippocampus. Recent experimental work has found that salient but task-irrelevant stimuli are systematically excluded from these replay epochs, suggesting that replay samples from an abstracted model of the world, rather than verbatim previous experiences. We find that this phenomenon can be explained parsimoniously and biologically plausibly by a Hebbian spike time-dependent plasticity rule at inhibitory synapses. Using spiking networks at three levels of abstraction–leaky integrate-and-fire, biophysically detailed, and abstract binary–we show that this rule enables efficient inference of a model of the structure of the world. While plasticity has previously mainly been studied at excitatory synapses, we find that plasticity at excitatory synapses alone is insufficient to accomplish this type of structural learning. We present theoretical results in a simplified model showing that in the presence of Hebbian excitatory and inhibitory plasticity, the replayed sequences form a statistical estimator of a latent sequence, which converges asymptotically to the ground truth. Our work outlines a direct link between the synaptic and cognitive levels of memory consolidation, and highlights a potential conceptually distinct role for inhibition in computing with SNNs.

Merging insights from artificial and biological neural networks for neuromorphic intelligence

Training Dynamic Spiking Neural Network via Forward Propagation Through Time

With recent advances in learning algorithms, recurrent networks of spiking neurons are achieving performance competitive with standard recurrent neural networks. Still, these learning algorithms are limited to small networks of simple spiking neurons and modest-length temporal sequences, as they impose high memory requirements, have difficulty training complex neuron models, and are incompatible with online learning.Taking inspiration from the concept of Liquid Time-Constant (LTCs), we introduce a novel class of spiking neurons, the Liquid Time-Constant Spiking Neuron (LTC-SN), resulting in functionality similar to the gating operation in LSTMs. We integrate these neurons in SNNs that are trained with FPTT and demonstrate that thus trained LTC-SNNs outperform various SNNs trained with BPTT on long sequences while enabling online learning and drastically reducing memory complexity. We show this for several classical benchmarks that can easily be varied in sequence length, like the Add Task and the DVS-gesture benchmark. We also show how FPTT-trained LTC-SNNs can be applied to large convolutional SNNs, where we demonstrate novel state-of-the-art for online learning in SNNs on a number of standard benchmarks (S-MNIST, R-MNIST, DVS-GESTURE) and also show that large feedforward SNNs can be trained successfully in an online manner to near (Fashion-MNIST, DVS-CIFAR10) or exceeding (PS-MNIST, R-MNIST) state-of-the-art performance as obtained with offline BPTT. Finally, the training and memory efficiency of FPTT enables us to directly train SNNs in an end-to-end manner at network sizes and complexity that was previously infeasible: we demonstrate this by training in an end-to-end fashion the first deep and performant spiking neural network for object localization and recognition. Taken together, we out contribution enable for the first time training large-scale complex spiking neural network architectures online and on long temporal sequences.

Universal function approximation in balanced spiking networks through convex-concave boundary composition

The spike-threshold nonlinearity is a fundamental, yet enigmatic, component of biological computation — despite its role in many theories, it has evaded definitive characterisation. Indeed, much classic work has attempted to limit the focus on spiking by smoothing over the spike threshold or by approximating spiking dynamics with firing-rate dynamics. Here, we take a novel perspective that captures the full potential of spike-based computation. Based on previous studies of the geometry of efficient spike-coding networks, we consider a population of neurons with low-rank connectivity, allowing us to cast each neuron’s threshold as a boundary in a space of population modes, or latent variables. Each neuron divides this latent space into subthreshold and suprathreshold areas. We then demonstrate how a network of inhibitory (I) neurons forms a convex, attracting boundary in the latent coding space, and a network of excitatory (E) neurons forms a concave, repellant boundary. Finally, we show how the combination of the two yields stable dynamics at the crossing of the E and I boundaries, and can be mapped onto a constrained optimization problem. The resultant EI networks are balanced, inhibition-stabilized, and exhibit asynchronous irregular activity, thereby closely resembling cortical networks of the brain. Moreover, we demonstrate how such networks can be tuned to either suppress or amplify noise, and how the composition of inhibitory convex and excitatory concave boundaries can result in universal function approximation. Our work puts forth a new theory of biologically-plausible computation in balanced spiking networks, and could serve as a novel framework for scalable and interpretable computation with spikes.

Spiking Deep Learning with SpikingJelly

Behavioral Timescale Synaptic Plasticity (BTSP) for biologically plausible credit assignment across multiple layers via top-down gating of dendritic plasticity

A central problem in biological learning is how information about the outcome of a decision or behavior can be used to reliably guide learning across distributed neural circuits while obeying biological constraints. This “credit assignment” problem is commonly solved in artificial neural networks through supervised gradient descent and the backpropagation algorithm. In contrast, biological learning is typically modelled using unsupervised Hebbian learning rules. While these rules only use local information to update synaptic weights, and are sometimes combined with weight constraints to reflect a diversity of excitatory (only positive weights) and inhibitory (only negative weights) cell types, they do not prescribe a clear mechanism for how to coordinate learning across multiple layers and propagate error information accurately across the network. In recent years, several groups have drawn inspiration from the known dendritic non-linearities of pyramidal neurons to propose new learning rules and network architectures that enable biologically plausible multi-layer learning by processing error information in segregated dendrites. Meanwhile, recent experimental results from the hippocampus have revealed a new form of plasticity—Behavioral Timescale Synaptic Plasticity (BTSP)—in which large dendritic depolarizations rapidly reshape synaptic weights and stimulus selectivity with as little as a single stimulus presentation (“one-shot learning”). Here we explore the implications of this new learning rule through a biologically plausible implementation in a rate neuron network. We demonstrate that regulation of dendritic spiking and BTSP by top-down feedback signals can effectively coordinate plasticity across multiple network layers in a simple pattern recognition task. By analyzing hidden feature representations and weight trajectories during learning, we show the differences between networks trained with standard backpropagation, Hebbian learning rules, and BTSP.

Why dendrites matter for biological and artificial circuits

Algorithm-Hardware Co-design for Efficient and Robust Spiking Neural Networks

Introducing dendritic computations to SNNs with Dendrify

Current SNNs studies frequently ignore dendrites, the thin membranous extensions of biological neurons that receive and preprocess nearly all synaptic inputs in the brain. However, decades of experimental and theoretical research suggest that dendrites possess compelling computational capabilities that greatly influence neuronal and circuit functions. Notably, standard point-neuron networks cannot adequately capture most hallmark dendritic properties. Meanwhile, biophysically detailed neuron models are impractical for large-network simulations due to their complexity, and high computational cost. For this reason, we introduce Dendrify, a new theoretical framework combined with an open-source Python package (compatible with Brian2) that facilitates the development of bioinspired SNNs. Dendrify, through simple commands, can generate reduced compartmental neuron models with simplified yet biologically relevant dendritic and synaptic integrative properties. Such models strike a good balance between flexibility, performance, and biological accuracy, allowing us to explore dendritic contributions to network-level functions while paving the way for developing more realistic neuromorphic systems.

Optimization at the Single Neuron Level:​ Prediction of Spike Sequences and Emergence of Synaptic Plasticity Mechanisms

Intelligent behavior depends on the brain’s ability to anticipate future events. However, the learning rules that enable neurons to predict and fire ahead of sensory inputs remain largely unknown. We propose a plasticity rule based on pre-dictive processing, where the neuron learns a low-rank model of the synaptic input dynamics in its membrane potential. Neurons thereby amplify those synapses that maximally predict other synaptic inputs based on their temporal relations, which provide a solution to an optimization problem that can be implemented at the single-neuron level using only local information. Consequently, neurons learn sequences over long timescales and shift their spikes towards the first inputs in a sequence. We show that this mechanism can explain the development of anticipatory motion signaling and recall in the visual system. Furthermore, we demonstrate that the learning rule gives rise to several experimentally observed STDP (spike-timing-dependent plasticity) mechanisms. These findings suggest prediction as a guiding principle to orchestrate learning and synaptic plasticity in single neurons.

Norse: A library for gradient-based learning in Spiking Neural Networks

Norse aims to exploit the advantages of bio-inspired neural components, which are sparse and event-driven - a fundamental difference from artificial neural networks. Norse expands PyTorch with primitives for bio-inspired neural components, bringing you two advantages: a modern and proven infrastructure based on PyTorch and deep learning-compatible spiking neural network components.

Optimising spiking interneuron circuits for compartment-specific feedback

Cortical circuits process information by rich recurrent interactions between excitatory neurons and inhibitory interneurons. One of the prime functions of interneurons is to stabilize the circuit by feedback inhibition, but the level of specificity on which inhibitory feedback operates is not fully resolved. We hypothesized that inhibitory circuits could enable separate feedback control loops for different synaptic input streams, by means of specific feedback inhibition to different neuronal compartments. To investigate this hypothesis, we adopted an optimization approach. Leveraging recent advances in training spiking network models, we optimized the connectivity and short-term plasticity of interneuron circuits for compartment-specific feedback inhibition onto pyramidal neurons. Over the course of the optimization, the interneurons diversified into two classes that resembled parvalbumin (PV) and somatostatin (SST) expressing interneurons. The resulting circuit can be understood as a neural decoder that inverts the nonlinear biophysical computations performed within the pyramidal cells. Our model provides a proof of concept for studying structure-function relations in cortical circuits by a combination of gradient-based optimization and biologically plausible phenomenological models

StereoSpike: Depth Learning with a Spiking Neural Network

Depth estimation is an important computer vision task, useful in particular for navigation in autonomous vehicles, or for object manipulation in robotics. Here we solved it using an end-to-end neuromorphic approach, combining two event-based cameras and a Spiking Neural Network (SNN) with a slightly modified U-Net-like encoder-decoder architecture, that we named StereoSpike. More specifically, we used the Multi Vehicle Stereo Event Camera Dataset (MVSEC). It provides a depth ground-truth, which was used to train StereoSpike in a supervised manner, using surrogate gradient descent. We propose a novel readout paradigm to obtain a dense analog prediction –the depth of each pixel– from the spikes of the decoder. We demonstrate that this architecture generalizes very well, even better than its non-spiking counterparts, leading to state-of-the-art test accuracy. To the best of our knowledge, it is the first time that such a large-scale regression problem is solved by a fully spiking network. Finally, we show that low firing rates (<10%) can be obtained via regularization, with a minimal cost in accuracy. This means that StereoSpike could be implemented efficiently on neuromorphic chips, opening the door for low power real time embedded systems.

Unsupervised Spiking Neural Networks

Spiking Neural networks as Universal Function Approximators - SNUFA 2021

Like last year this online workshop brings together researchers in the field to present their work and discuss ways of translating these findings into a better understanding of neural circuits. Topics include artificial and biologically plausible learning algorithms and the dissection of trained spiking circuits toward understanding neural processing. We have a manageable number of talks with ample time for discussions. This year’s executive committee comprises Chiara Bartolozzi, Sander Bohté, Dan Goodman, and Friedemann Zenke.

Towards multipurpose biophysics-based mathematical models of cortical circuits

Starting with the work of Hodgkin and Huxley in the 1950s, we now have a fairly good understanding of how the spiking activity of neurons can be modelled mathematically. For cortical circuits the understanding is much more limited. Most network studies have considered stylized models with a single or a handful of neuronal populations consisting of identical neurons with statistically identical connection properties. However, real cortical networks have heterogeneous neural populations and much more structured synaptic connections. Unlike typical simplified cortical network models, real networks are also “multipurpose” in that they perform multiple functions. Historically the lack of computational resources has hampered the mathematical exploration of cortical networks. With the advent of modern supercomputers, however, simulations of networks comprising hundreds of thousands biologically detailed neurons are becoming feasible (Einevoll et al, Neuron, 2019). Further, a large-scale biologically network model of the mouse primary visual cortex comprising 230.000 neurons has recently been developed at the Allen Institute for Brain Science (Billeh et al, Neuron, 2020). Using this model as a starting point, I will discuss how we can move towards multipurpose models that incorporate the true biological complexity of cortical circuits and faithfully reproduce multiple experimental observables such as spiking activity, local field potentials or two-photon calcium imaging signals. Further, I will discuss how such validated comprehensive network models can be used to gain insights into the functioning of cortical circuits.

On temporal coding in spiking neural networks with alpha synaptic function

The timing of individual neuronal spikes is essential for biological brains to make fast responses to sensory stimuli. However, conventional artificial neural networks lack the intrinsic temporal coding ability present in biological networks. We propose a spiking neural network model that encodes information in the relative timing of individual neuron spikes. In classification tasks, the output of the network is indicated by the first neuron to spike in the output layer. This temporal coding scheme allows the supervised training of the network with backpropagation, using locally exact derivatives of the postsynaptic spike times with respect to presynaptic spike times. The network operates using a biologically-plausible alpha synaptic transfer function. Additionally, we use trainable synchronisation pulses that provide bias, add flexibility during training and exploit the decay part of the alpha function. We show that such networks can be trained successfully on noisy Boolean logic tasks and on the MNIST dataset encoded in time. The results show that the spiking neural network outperforms comparable spiking models on MNIST and achieves similar quality to fully connected conventional networks with the same architecture. We also find that the spiking network spontaneously discovers two operating regimes, mirroring the accuracy-speed trade-off observed in human decision-making: a slow regime, where a decision is taken after all hidden neurons have spiked and the accuracy is very high, and a fast regime, where a decision is taken very fast but the accuracy is lower. These results demonstrate the computational power of spiking networks with biological characteristics that encode information in the timing of individual neurons. By studying temporal coding in spiking networks, we aim to create building blocks towards energy-efficient and more complex biologically-inspired neural architectures.

Workshop on "Spiking neural networks as universal function approximators: Learning algorithms and applications

This is a two-day workshop. Sign up and see titles and abstracts on website.

Efficient cortical spike train decoding for brain-machine interface implants with recurrent spiking neural networks

Bernstein Conference 2024

Emergence of Synfire Chains in Functional Multi-Layer Spiking Neural Networks

Bernstein Conference 2024

A feedback control algorithm for online learning in Spiking Neural Networks and Neuromorphic devices

Bernstein Conference 2024

Enhancing learning through neuromodulation-aware spiking neural networks

Bernstein Conference 2024

Experiment-based Models to Study Local Learning Rules for Spiking Neural Networks

Bernstein Conference 2024

Parameter specification in spiking neural networks using simulation-based inference

Bernstein Conference 2024

On The Role Of Temporal Hierarchy In Spiking Neural Networks

Bernstein Conference 2024

Seamless Deployment of Pre-trained Spiking Neural Networks onto SpiNNaker2

Bernstein Conference 2024

Smooth exact gradient descent learning in spiking neural networks

Bernstein Conference 2024

Using Dynamical Systems Theory to Improve Temporal Credit Assignment in Spiking Neural Networks

Bernstein Conference 2024

How spiking neural networks can flexibly trade off performance and energy use

COSYNE 2022

How spiking neural networks can flexibly trade off performance and energy use

COSYNE 2022

Supervised learning and interpretation of plasticity rules in spiking neural networks

COSYNE 2022

Supervised learning and interpretation of plasticity rules in spiking neural networks

COSYNE 2022

Effects of Neural Heterogeneity on the Low-Dimensional Dynamics of Spiking Neural Networks

COSYNE 2023

Hierarchical structure of combinatorial code optimizes representation in spiking neural networks.

COSYNE 2025

A mechanism for selective attention in biophysically realistic Daleian spiking neural networks

COSYNE 2025

Fast gradient-free activation maximization for neurons in spiking neural networks

FENS Forum 2024

Library of Dynamics: Linking parameters and behaviour of spiking neural networks

FENS Forum 2024

Mechanisms of attention in biophysiologically realistic Daleian spiking neural networks

FENS Forum 2024

Population geometry enables fast sampling in spiking neural networks

Neuromatch 5