Effective Surrogate Gradient Learning With High-Order Information Bottleneck for Spike-Based Machine Intelligence

Temporal resolution of spike coding in feedforward networks with signal convergence and divergence

Convergent and divergent structures in the networks that make up biological brains are found across many species and brain regions at various spatial scales. Neurons in these networks fire action potentials, or "spikes," whose precise timing is becoming increasingly appreciated as large sources of information about both sensory input and motor output. In this work, we investigate the extent to which feedforward convergent/divergent network structure is related to the gain in information of spike timing representations over spike count representations. While previous theories on coding in convergent and divergent networks have largely neglected the role of precise spike timing, our model and analyses place this aspect at the forefront. For a suite of stimuli with different timescales, we demonstrate that structural bottlenecks-small groups of neurons post-synaptic to network convergence-have a stronger preference for spike timing codes than expansion layers created by structural divergence. We further show that this relationship can be generalized across different spike-generating models and measures of coding capacity, implying a potentially fundamental link between network structure and coding strategy using spikes. Additionally, we found that a simple network model based on convergence and divergence ratios of a hawkmoth (Manduca sexta) nervous system can reproduce the relative contribution of spike timing information in its motor output, providing testable predictions on optimal temporal resolutions of spike coding across the moth sensory-motor pathway at both the single-neuron and population levels.

Exploring spiking neural networks for deep reinforcement learning in robotic tasks

Spiking Neural Networks (SNNs) stand as the third generation of Artificial Neural Networks (ANNs), mirroring the functionality of the mammalian brain more closely than their predecessors. Their computational units, spiking neurons, characterized by Ordinary Differential Equations (ODEs), allow for dynamic system representation, with spikes serving as the medium for asynchronous communication among neurons. Due to their inherent ability to capture input dynamics, SNNs hold great promise for deep networks in Reinforcement Learning (RL) tasks. Deep RL (DRL), and in particular Proximal Policy Optimization (PPO) has been proven to be valuable for training robots due to the difficulty in creating comprehensive offline datasets that capture all environmental features. DRL combined with SNNs offers a compelling solution for tasks characterized by temporal complexity. In this work, we study the effectiveness of SNNs on DRL tasks leveraging a novel framework we developed for training SNNs with PPO in the Isaac Gym simulator implemented using the skrl library. Thanks to its significantly faster training speed compared to available SNN DRL tools, the framework allowed us to: (i) Perform an effective exploration of SNN configurations for DRL robotic tasks; (ii) Compare SNNs and ANNs for various network configurations such as the number of layers and neurons. Our work demonstrates that in DRL tasks the optimal SNN topology has a lower number of layers than ANN and we highlight how the state-of-art SNN architectures used in complex RL tasks, such as Ant, SNNs have difficulties fully leveraging deeper layers. Finally, we applied the best topology identified thanks to our Isaac Gym-based framework on Ant-v4 benchmark running on MuJoCo simulator, exhibiting a performance improvement by a factor of 4.4 × over the state-of-art SNN trained on the same task.

Decoding the application of deep learning in neuroscience: a bibliometric analysis

The application of deep learning in neuroscience holds unprecedented potential for unraveling the complex dynamics of the brain. Our bibliometric analysis, spanning from 2012 to 2023, delves into the integration of deep learning in neuroscience, shedding light on the evolutionary trends and identifying pivotal research hotspots. Through the examination of 421 articles, this study unveils a significant growth in interdisciplinary research, marked by the burgeoning application of deep learning techniques in understanding neural mechanisms and addressing neurological disorders. Central to our findings is the critical role of classification algorithms, models, and neural networks in advancing neuroscience, highlighting their efficacy in interpreting complex neural data, simulating brain functions, and translating theoretical insights into practical diagnostics and therapeutic interventions. Additionally, our analysis delineates a thematic evolution, showcasing a shift from foundational methodologies toward more specialized and nuanced approaches, particularly in areas like EEG analysis and convolutional neural networks. This evolution reflects the field's maturation and its adaptation to technological advancements. The study further emphasizes the importance of interdisciplinary collaborations and the adoption of cutting-edge technologies to foster innovation in decoding the cerebral code. The current study provides a strategic roadmap for future explorations, urging the scientific community toward areas ripe for breakthrough discoveries and practical applications. This analysis not only charts the past and present landscape of deep learning in neuroscience but also illuminates pathways for future research, underscoring the transformative impact of deep learning on our understanding of the brain.

Brain-inspired biomimetic robot control: a review

Complex robotic systems, such as humanoid robot hands, soft robots, and walking robots, pose a challenging control problem due to their high dimensionality and heavy non-linearities. Conventional model-based feedback controllers demonstrate robustness and stability but struggle to cope with the escalating system design and tuning complexity accompanying larger dimensions. In contrast, data-driven methods such as artificial neural networks excel at representing high-dimensional data but lack robustness, generalization, and real-time adaptiveness. In response to these challenges, researchers are directing their focus to biological paradigms, drawing inspiration from the remarkable control capabilities inherent in the human body. This has motivated the exploration of new control methods aimed at closely emulating the motor functions of the brain given the current insights in neuroscience. Recent investigation into these Brain-Inspired control techniques have yielded promising results, notably in tasks involving trajectory tracking and robot locomotion. This paper presents a comprehensive review of the foremost trends in biomimetic brain-inspired control methods to tackle the intricacies associated with controlling complex robotic systems.

A lightweight data-driven spiking neuronal network model of Drosophila olfactory nervous system with dedicated hardware support

Data-driven spiking neuronal network (SNN) models enable in-silico analysis of the nervous system at the cellular and synaptic level. Therefore, they are a key tool for elucidating the information processing principles of the brain. While extensive research has focused on developing data-driven SNN models for mammalian brains, their complexity poses challenges in achieving precision. Network topology often relies on statistical inference, and the functions of specific brain regions and supporting neuronal activities remain unclear. Additionally, these models demand huge computing facilities and their simulation speed is considerably slower than real-time. Here, we propose a lightweight data-driven SNN model that strikes a balance between simplicity and reproducibility. The model is built using a qualitative modeling approach that can reproduce key dynamics of neuronal activity. We target the Drosophila olfactory nervous system, extracting its network topology from connectome data. The model was successfully implemented on a small entry-level field-programmable gate array and simulated the activity of a network in real-time. In addition, the model reproduced olfactory associative learning, the primary function of the olfactory system, and characteristic spiking activities of different neuron types. In sum, this paper propose a method for building data-driven SNN models from biological data. Our approach reproduces the function and neuronal activities of the nervous system and is lightweight, acceleratable with dedicated hardware, making it scalable to large-scale networks. Therefore, our approach is expected to play an important role in elucidating the brain's information processing at the cellular and synaptic level through an analysis-by-construction approach. In addition, it may be applicable to edge artificial intelligence systems in the future.

Information bottleneck-based Hebbian learning rule naturally ties working memory and synaptic updates

Deep neural feedforward networks are effective models for a wide array of problems, but training and deploying such networks presents a significant energy cost. Spiking neural networks (SNNs), which are modeled after biologically realistic neurons, offer a potential solution when deployed correctly on neuromorphic computing hardware. Still, many applications train SNNs offline, and running network training directly on neuromorphic hardware is an ongoing research problem. The primary hurdle is that back-propagation, which makes training such artificial deep networks possible, is biologically implausible. Neuroscientists are uncertain about how the brain would propagate a precise error signal backward through a network of neurons. Recent progress addresses part of this question, e.g., the weight transport problem, but a complete solution remains intangible. In contrast, novel learning rules based on the information bottleneck (IB) train each layer of a network independently, circumventing the need to propagate errors across layers. Instead, propagation is implicit due the layers' feedforward connectivity. These rules take the form of a three-factor Hebbian update a global error signal modulates local synaptic updates within each layer. Unfortunately, the global signal for a given layer requires processing multiple samples concurrently, and the brain only sees a single sample at a time. We propose a new three-factor update rule where the global signal correctly captures information across samples via an auxiliary memory network. The auxiliary network can be trained a priori independently of the dataset being used with the primary network. We demonstrate comparable performance to baselines on image classification tasks. Interestingly, unlike back-propagation-like schemes where there is no link between learning and memory, our rule presents a direct connection between working memory and synaptic updates. To the best of our knowledge, this is the first rule to make this link explicit. We explore these implications in initial experiments examining the effect of memory capacity on learning performance. Moving forward, this work suggests an alternate view of learning where each layer balances memory-informed compression against task performance. This view naturally encompasses several key aspects of neural computation, including memory, efficiency, and locality.

Robust compression and detection of epileptiform patterns in ECoG using a real-time spiking neural network hardware framework

Interictal Epileptiform Discharges (IED) and High Frequency Oscillations (HFO) in intraoperative electrocorticography (ECoG) may guide the surgeon by delineating the epileptogenic zone. We designed a modular spiking neural network (SNN) in a mixed-signal neuromorphic device to process the ECoG in real-time. We exploit the variability of the inhomogeneous silicon neurons to achieve efficient sparse and decorrelated temporal signal encoding. We interface the full-custom SNN device to the BCI2000 real-time framework and configure the setup to detect HFO and IED co-occurring with HFO (IED-HFO). We validate the setup on pre-recorded data and obtain HFO rates that are concordant with a previously validated offline algorithm (Spearman's ρ = 0.75, p = 1e-4), achieving the same postsurgical seizure freedom predictions for all patients. In a remote on-line analysis, intraoperative ECoG recorded in Utrecht was compressed and transferred to Zurich for SNN processing and successful IED-HFO detection in real-time. These results further demonstrate how automated remote real-time detection may enable the use of HFO in clinical practice.

SLoN: a spiking looming perception network exploiting neural encoding and processing in ON/OFF channels

Looming perception, the ability to sense approaching objects, is crucial for the survival of humans and animals. After hundreds of millions of years of evolutionary development, biological entities have evolved efficient and robust looming perception visual systems. However, current artificial vision systems fall short of such capabilities. In this study, we propose a novel spiking neural network for looming perception that mimics biological vision to communicate motion information through action potentials or spikes, providing a more realistic approach than previous artificial neural networks based on sum-then-activate operations. The proposed spiking looming perception network (SLoN) comprises three core components. Neural encoding, known as phase coding, transforms video signals into spike trains, introducing the concept of phase delay to depict the spatial-temporal competition between phasic excitatory and inhibitory signals shaping looming selectivity. To align with biological substrates where visual signals are bifurcated into parallel ON/OFF channels encoding brightness increments and decrements separately to achieve specific selectivity to ON/OFF-contrast stimuli, we implement eccentric down-sampling at the entrance of ON/OFF channels, mimicking the foveal region of the mammalian receptive field with higher acuity to motion, computationally modeled with a leaky integrate-and-fire (LIF) neuronal network. The SLoN model is deliberately tested under various visual collision scenarios, ranging from synthetic to real-world stimuli. A notable achievement is that the SLoN selectively spikes for looming features concealed in visual streams against other categories of movements, including translating, receding, grating, and near misses, demonstrating robust selectivity in line with biological principles. Additionally, the efficacy of the ON/OFF channels, the phase coding with delay, and the eccentric visual processing are further investigated to demonstrate their effectiveness in looming perception. The cornerstone of this study rests upon showcasing a new paradigm for looming perception that is more biologically plausible in light of biological motion perception.

Rethinking skip connections in Spiking Neural Networks with Time-To-First-Spike coding

Time-To-First-Spike (TTFS) coding in Spiking Neural Networks (SNNs) offers significant advantages in terms of energy efficiency, closely mimicking the behavior of biological neurons. In this work, we delve into the role of skip connections, a widely used concept in Artificial Neural Networks (ANNs), within the domain of SNNs with TTFS coding. Our focus is on two distinct types of skip connection architectures: (1) addition-based skip connections, and (2) concatenation-based skip connections. We find that addition-based skip connections introduce an additional delay in terms of spike timing. On the other hand, concatenation-based skip connections circumvent this delay but produce time gaps between after-convolution and skip connection paths, thereby restricting the effective mixing of information from these two paths. To mitigate these issues, we propose a novel approach involving a learnable delay for skip connections in the concatenation-based skip connection architecture. This approach successfully bridges the time gap between the convolutional and skip branches, facilitating improved information mixing. We conduct experiments on public datasets including MNIST and Fashion-MNIST, illustrating the advantage of the skip connection in TTFS coding architectures. Additionally, we demonstrate the applicability of TTFS coding on beyond image recognition tasks and extend it to scientific machine-learning tasks, broadening the potential uses of SNNs.

Bidirectional feature pyramid attention-based temporal convolutional network model for motor imagery electroencephalogram classification

Introduction

As an interactive method gaining popularity, brain-computer interfaces (BCIs) aim to facilitate communication between the brain and external devices. Among the various research topics in BCIs, the classification of motor imagery using electroencephalography (EEG) signals has the potential to greatly improve the quality of life for people with disabilities.

Methods

This technology assists them in controlling computers or other devices like prosthetic limbs, wheelchairs, and drones. However, the current performance of EEG signal decoding is not sufficient for real-world applications based on Motor Imagery EEG (MI-EEG). To address this issue, this study proposes an attention-based bidirectional feature pyramid temporal convolutional network model for the classification task of MI-EEG. The model incorporates a multi-head self-attention mechanism to weigh significant features in the MI-EEG signals. It also utilizes a temporal convolution network (TCN) to separate high-level temporal features. The signals are enhanced using the sliding-window technique, and channel and time-domain information of the MI-EEG signals is extracted through convolution.

Results

Additionally, a bidirectional feature pyramid structure is employed to implement attention mechanisms across different scales and multiple frequency bands of the MI-EEG signals. The performance of our model is evaluated on the BCI Competition IV-2a dataset and the BCI Competition IV-2b dataset, and the results showed that our model outperformed the state-of-the-art baseline model, with an accuracy of 87.5 and 86.3% for the subject-dependent, respectively.

Discussion

In conclusion, the BFATCNet model offers a novel approach for EEG-based motor imagery classification in BCIs, effectively capturing relevant features through attention mechanisms and temporal convolutional networks. Its superior performance on the BCI Competition IV-2a and IV-2b datasets highlights its potential for real-world applications. However, its performance on other datasets may vary, necessitating further research on data augmentation techniques and integration with multiple modalities to enhance interpretability and generalization. Additionally, reducing computational complexity for real-time applications is an important area for future work.