Towards efficient keyword spotting using spike-based time difference encoders

Keyword spotting in edge devices is becoming increasingly important as voice-activated assistants are widely used. However, its deployment is often limited by the extreme low-power constraints of the target embedded systems. Here, we explore the Temporal Difference Encoder (TDE) performance in keyword spotting. This recent neuron model encodes the time difference in instantaneous frequency and spike count to perform efficient keyword spotting with neuromorphic processors. We use the TIdigits dataset of spoken digits with a formant decomposition and rate-based encoding into spikes. We compare three Spiking Neural Networks (SNNs) architectures to learn and classify spatio-temporal signals. The proposed SNN architectures are made of three layers with variation in its hidden layer composed of either (1) feedforward TDE, (2) feedforward Current-Based Leaky Integrate-and-Fire (CuBa-LIF), or (3) recurrent CuBa-LIF neurons. We first show that the spike trains of the frequency-converted spoken digits have a large amount of information in the temporal domain, reinforcing the importance of better exploiting temporal encoding for such a task. We then train the three SNNs with the same number of synaptic weights to quantify and compare their performance based on the accuracy and synaptic operations. The resulting accuracy of the feedforward TDE network (89%) is higher than the feedforward CuBa-LIF network (71%) and close to the recurrent CuBa-LIF network (91%). However, the feedforward TDE-based network performs 92% fewer synaptic operations than the recurrent CuBa-LIF network with the same amount of synapses. In addition, the results of the TDE network are highly interpretable and correlated with the frequency and timescale features of the spoken keywords in the dataset. Our findings suggest that the TDE is a promising neuron model for scalable event-driven processing of spatio-temporal patterns.

TDE-3: An improved prior for optical flow computation in spiking neural networks

Motion detection is a primary task required for robotic systems to perceive and navigate in their environment. Proposed in the literature bioinspired neuromorphic Time-Difference Encoder (TDE-2) combines event-based sensors and processors with spiking neural networks to provide real-time and energy-efficient motion detection through extracting temporal correlations between two points in space. However, on the algorithmic level, this design leads to loss of direction-selectivity of individual TDEs in textured environments. Here we propose an augmented 3-point TDE (TDE-3) with additional inhibitory input that makes TDE-3 direction-selectivity robust in textured environments. We developed a procedure to train the new TDE-3 using backpropagation through time and surrogate gradients to linearly map input velocities into an output spike count or an Inter-Spike Interval (ISI). Our work is the first instance of training a spiking neuron to have a specific ISI. Using synthetic data we compared training and inference with spike count and ISI with respect to changes in stimuli dynamic range, spatial frequency, and level of noise. ISI turns out to be more robust towards variation in spatial frequency, whereas the spike count is a more reliable training signal in the presence of noise. We performed the first in-depth quantitative investigation of optical flow coding with TDE and compared TDE-2 vs TDE-3 in terms of energy-efficiency and coding precision. Results show that on the network level both detectors show similar precision (20 degree angular error, 88% correlation with ground truth). Yet, due to the more robust direction-selectivity of individual TDEs, TDE-3 based network spike less and hence is more energy-efficient. Reported precision is on par with model-based methods but the spike-based processing of the TDEs provides allows more energy-efficient inference with neuromorphic hardware.

Low-power event-based face detection with asynchronous neuromorphic hardware

The rise of mobility, IoT and wearables has shifted processing to the edge of the sensors, driven by the need to reduce latency, communication costs and overall energy consumption. While deep learning models have achieved remarkable results in various domains, their deployment at the edge for real-time applications remains computationally expensive. Neuromorphic computing emerges as a promising paradigm shift, characterized by co-localized memory and computing as well as event-driven asynchronous sensing and processing. In this work, we demonstrate the possibility of solving the ubiquitous computer vision task of object detection at the edge with low-power requirements, using the event-based N-Caltech101 dataset. We present the first instance of an on-chip spiking neural network for event-based face detection deployed on the SynSense Speck neuromorphic chip, which comprises both an event-based sensor and a spike-based asynchronous processor implementing Integrate-and-Fire neurons. We show how to reduce precision discrepancies between off-chip clock-driven simulation used for training and on-chip event-driven inference. This involves using a multi-spike version of the Integrate-and-Fire neuron on simulation, where spikes carry values that are proportional to the extent the membrane potential exceeds the firing threshold. We propose a robust strategy to train spiking neural networks with back-propagation through time using multi-spike activation and firing rate regularization and demonstrate how to decode output spikes into bounding boxes. We show that the power consumption of the chip is directly proportional to the number of synaptic operations in the spiking neural network, and we explore the trade-off between power consumption and detection precision with different firing rate regularization, achieving an on-chip face detection mAP[0.5] of ~0.6 while consuming only ~20 mW.

NeuroBench: Advancing Neuromorphic Computing through Collaborative, Fair and Representative Benchmarking

The field of neuromorphic computing holds great promise in terms of advancing computing efficiency and capabilities by following brain-inspired principles. However, the rich diversity of techniques employed in neuromorphic research has resulted in a lack of clear standards for benchmarking, hindering effective evaluation of the advantages and strengths of neuromorphic methods compared to traditional deep-learning-based methods. This paper presents a collaborative effort, bringing together members from academia and the industry, to define benchmarks for neuromorphic computing: NeuroBench. The goals of NeuroBench are to be a collaborative, fair, and representative benchmark suite developed by the community, for the community. In this paper, we discuss the challenges associated with benchmarking neuromorphic solutions, and outline the key features of NeuroBench. We believe that NeuroBench will be a significant step towards defining standards that can unify the goals of neuromorphic computing and drive its technological progress. Please visit neurobench.ai for the latest updates on the benchmark tasks and metrics.

ETLP: Event-based Three-factor Local Plasticity for online learning with neuromorphic hardware

Neuromorphic perception with event-based sensors, asynchronous hardware and spiking neurons is showing promising results for real-time and energy-efficient inference in embedded systems. The next promise of brain-inspired computing is to enable adaptation to changes at the edge with online learning. However, the parallel and distributed architectures of neuromorphic hardware based on co-localized compute and memory imposes locality constraints to the on-chip learning rules. We propose in this work the Event-based Three-factor Local Plasticity (ETLP) rule that uses (1) the pre-synaptic spike trace, (2) the post-synaptic membrane voltage and (3) a third factor in the form of projected labels with no error calculation, that also serve as update triggers. We apply ETLP with feedforward and recurrent spiking neural networks on visual and auditory event-based pattern recognition, and compare it to Back-Propagation Through Time (BPTT) and eProp. We show a competitive performance in accuracy with a clear advantage in the computational complexity for ETLP. We also show that when using local plasticity, threshold adaptation in spiking neurons and a recurrent topology are necessary to learn spatio-temporal patterns with a rich temporal structure. Finally, we provide a proof of concept hardware implementation of ETLP on FPGA to highlight the simplicity of its computational primitives and how they can be mapped into neuromorphic hardware for online learning with low-energy consumption and real-time interaction.

A Comparison of Temporal Encoders for Neuromorphic Keyword Spotting with Few Neurons

With the expansion of AI-powered virtual assistants, there is a need for low-power keyword spotting systems providing a "wake-up" mechanism for subsequent computationally expensive speech recognition. One promising approach is the use of neuromorphic sensors and spiking neural networks (SNNs) implemented in neuromorphic processors for sparse event-driven sensing. However, this requires resource-efficient SNN mechanisms for temporal encoding, which need to consider that these systems process information in a streaming manner, with physical time being an intrinsic property of their operation. In this work, two candidate neurocomputational elements for temporal encoding and feature extraction in SNNs described in recent literature - the spiking time-difference encoder (TDE) and disynaptic excitatory-inhibitory (E-I) elements - are comparatively investigated in a keyword-spotting task on formants computed from spoken digits in the TIDIGITS dataset. While both encoders improve performance over direct classification of the formant features in the training data, enabling a complete binary classification with a logistic regression model, they show no clear improvements on the test set. Resource-efficient keyword spotting applications may benefit from the use of these encoders, but further work on methods for learning the time constants and weights is required to investigate their full potential.

Impact of spiking neurons leakages and network recurrences on event-based spatio-temporal pattern recognition

Spiking neural networks coupled with neuromorphic hardware and event-based sensors are getting increased interest for low-latency and low-power inference at the edge. However, multiple spiking neuron models have been proposed in the literature with different levels of biological plausibility and different computational features and complexities. Consequently, there is a need to define the right level of abstraction from biology in order to get the best performance in accurate, efficient and fast inference in neuromorphic hardware. In this context, we explore the impact of synaptic and membrane leakages in spiking neurons. We confront three neural models with different computational complexities using feedforward and recurrent topologies for event-based visual and auditory pattern recognition. Our results show that, in terms of accuracy, leakages are important when there are both temporal information in the data and explicit recurrence in the network. In addition, leakages do not necessarily increase the sparsity of spikes flowing in the network. We also investigate the impact of heterogeneity in the time constant of leakages, and the results show a slight improvement in accuracy when using data with a rich temporal structure. These results advance our understanding of the computational role of the neural leakages and network recurrences, and provide valuable insights for the design of compact and energy-efficient neuromorphic hardware for embedded systems.

Spike-based local synaptic plasticity: A survey of computational models and neuromorphic circuits

Understanding how biological neural networks carry out learning using spike-based local plasticity mechanisms can lead to the development of powerful, energy-efficient, and adaptive neuromorphic processing systems. A large number of spike-based learning models have recently been proposed following different approaches. However, it is difficult to assess if and how they could be mapped onto neuromorphic hardware, and to compare their features and ease of implementation. To this end, in this survey, we provide a comprehensive overview of representative brain-inspired synaptic plasticity models and mixed-signal \acs{CMOS} neuromorphic circuits within a unified framework. We review historical, bottom-up, and top-down approaches to modeling synaptic plasticity, and we identify computational primitives that can support low-latency and low-power hardware implementations of spike-based learning rules. We provide a common definition of a locality principle based on pre- and post-synaptic neuron information, which we propose as a fundamental requirement for physical implementations of synaptic plasticity. Based on this principle, we compare the properties of these models within the same framework, and describe the mixed-signal electronic circuits that implement their computing primitives, pointing out how these building blocks enable efficient on-chip and online learning in neuromorphic processing systems.

Braille Letter Reading: A Benchmark for Spatio-Temporal Pattern Recognition on Neuromorphic Hardware

Spatio-temporal pattern recognition is a fundamental ability of the brain which is required for numerous real-world applications. Recent deep learning approaches have reached outstanding accuracy in such tasks, but their implementation on conventional embedded solutions is still very computationally and energy expensive. Tactile sensing in robotic applications is a representative example where real-time processing and energy-efficiency are required. Following a brain-inspired computing approach, we propose a new benchmark for spatio-temporal tactile pattern recognition at the edge through braille letters reading. We recorded a new braille letters dataset based on the capacitive tactile sensors/fingertip of the iCub robot, then we investigated the importance of temporal information and the impact of event-based encoding for spike-based/event-based computation. Afterwards, we trained and compared feed-forward and recurrent spiking neural networks (SNNs) offline using back-propagation through time with surrogate gradients, then we deployed them on the Intel Loihi neuromorphic chip for fast and efficient inference. We confronted our approach to standard classifiers, in particular to a Long Short-Term Memory (LSTM) deployed on the embedded Nvidia Jetson GPU in terms of classification accuracy, power/energy consumption and computational delay. Our results show that the LSTM outperforms the recurrent SNN in terms of accuracy by 14%. However, the recurrent SNN on Loihi is 237 times more energy-efficient than the LSTM on Jetson, requiring an average power of only 31mW. This work proposes a new benchmark for tactile sensing and highlights the challenges and opportunities of event-based encoding, neuromorphic hardware and spike-based computing for spatio-temporal pattern recognition at the edge.

A unified software/hardware scalable architecture for brain-inspired computing based on self-organizing neural models

The field of artificial intelligence has significantly advanced over the past decades, inspired by discoveries from the fields of biology and neuroscience. The idea of this work is inspired by the process of self-organization of cortical areas in the human brain from both afferent and lateral/internal connections. In this work, we develop an original brain-inspired neural model associating Self-Organizing Maps (SOM) and Hebbian learning in the Reentrant SOM (ReSOM) model. The framework is applied to multimodal classification problems. Compared to existing methods based on unsupervised learning with post-labeling, the model enhances the state-of-the-art results. This work also demonstrates the distributed and scalable nature of the model through both simulation results and hardware execution on a dedicated FPGA-based platform named SCALP (Self-configurable 3D Cellular Adaptive Platform). SCALP boards can be interconnected in a modular way to support the structure of the neural model. Such a unified software and hardware approach enables the processing to be scaled and allows information from several modalities to be merged dynamically. The deployment on hardware boards provides performance results of parallel execution on several devices, with the communication between each board through dedicated serial links. The proposed unified architecture, composed of the ReSOM model and the SCALP hardware platform, demonstrates a significant increase in accuracy thanks to multimodal association, and a good trade-off between latency and power consumption compared to a centralized GPU implementation.