Continuous learning of spiking networks trained with local rules

Antonov, Dmitry; Sviatov, Kirill; Sukhov, Sergey

doi:10.1016/j.neunet.2022.09.003

Cited by 5 publications

(10 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As for other SNNs-based methods for continual learning, (Antonov, Sviatov, and Sukhov 2022) determines the importance of synaptic weights via stochastic Langevin dynamics with local STDP and achieved continual learning by unsupervised learning. (Skatchkovsky, Jang, and Simeone 2022) introduced an online rule base on the Bayesian SNN model.…”

Section: Related Workmentioning

confidence: 99%

Efficient Spiking Neural Networks with Sparse Selective Activation for Continual Learning

Shen,

Ni,

et al. 2024

AAAI

View full text Add to dashboard Cite

The next generation of machine intelligence requires the capability of continual learning to acquire new knowledge without forgetting the old one while conserving limited computing resources. Spiking neural networks (SNNs), compared to artificial neural networks (ANNs), have more characteristics that align with biological neurons, which may be helpful as a potential gating function for knowledge maintenance in neural networks. Inspired by the selective sparse activation principle of context gating in biological systems, we present a novel SNN model with selective activation to achieve continual learning. The trace-based K-Winner-Take-All (K-WTA) and variable threshold components are designed to form the sparsity in selective activation in spatial and temporal dimensions of spiking neurons, which promotes the subpopulation of neuron activation to perform specific tasks. As a result, continual learning can be maintained by routing different tasks via different populations of neurons in the network. The experiments are conducted on MNIST and CIFAR10 datasets under the class incremental setting. The results show that the proposed SNN model achieves competitive performance similar to and even surpasses the other regularization-based methods deployed under traditional ANNs.

show abstract

Section: Related Workmentioning

confidence: 99%

Efficient Spiking Neural Networks with Sparse Selective Activation for Continual Learning

Shen,

Ni,

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Input-output relationship of a push-pull LIF neuron pair in the average spike rate domain. Simulated using a single spiking input connected with a weight of 1 with illustration parameters τ s = 0.01s, µ = 0.15, τ m = 0.002s in (11), (13).…”

Section: Spiking Neural Network and Stdp Learningmentioning

confidence: 99%

“…In recent years, the remarkable ability of biological entities for continual learning has triggered research in the emerging field of neuromorphic continual learning, which seeks to study how bio-plausible neural network architectures such as Hebbian learning networks and Spiking Neural Networks (SNNs) equipped with Spike-Timing-Dependent Plasticity (STDP) can be used for designing continual-learning systems without suffering from the compute-and memory-intensive overheads of traditional Deep Learning systems [9,10,11,12,13,14,15]. Following these realisations, a number of works covering a wide range of applications have been proposed, from task-and classincremental learning setups [9] to the continual learning of object detection and robot navigation [6,8].…”

Section: Introductionmentioning

confidence: 99%

Event Camera Data Classification Using Spiking Networks with Spike-Timing-Dependent Plasticity

Safa

Ocket

Bourdoux

et al. 2022

2022 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

We present an optimization-based theory describing spiking cortical ensembles equipped with Spike-Timing-Dependent Plasticity (STDP) learning, as empirically observed in the visual cortex. Using this generic framework, we build a class of global, action-based and convolutional feature descriptors for event-based cameras that we assess on the N-MNIST, the CIFAR10-DVS and the IBM DVS128 Gesture datasets. We report significant accuracy improvements compared to conventional state-of-the-art event-based feature descriptors (+8.2% on CIFAR10-DVS) and compared to state-of-the-art STDP-based systems (+9.3% on N-MNIST, +7.74% on IBM DVS128 Gesture). In addition to ultra-low-power learning in neuromorphic edge devices, our work contributes towards a biologically-plausible, optimization-based theory of cortical vision.

show abstract

“…However, weight training in DNNs typically relies on the backpropagation algorithm, which is a powerful method but energy-intensive and sensitive to non-idealities in 2 the weight updates of memristors [29,31,32]. Moreover, the biological plausibility of backpropagation remains an open question [33].…”

Section: Introductionmentioning

confidence: 99%

Actor-Critic Networks with Analogue Memristors Mimicking Reward-Based Learning

Portner,

Zellweger,

Martinelli

et al. 2024

Preprint

View full text Add to dashboard Cite

Advancements in memristive devices have given rise to a new generation of specialized hardware for bio-inspired computing. However, the majority of these implementations only draw partial inspiration from the architecture and functionalities of the mammalian brain. Moreover, the use of memristive hardware is typically restricted to specific elements within the learning algorithm, leaving computationally expensive operations to be executed in software. Here, we demonstrate actor-critic temporal difference (TD) learning on analogue memristors, mirroring the principles of reward-based learning in a neural network architecture similar to the one found in biology. Within the learning algorithm, memristors are used as multi-purpose elements: They act as synaptic weights that are trained online, they calculate the weight updates directly in hardware, and they compute the actions for navigating through the environment. Thanks to these features, weight training can take place entirely in-memory, eliminating the need for data movement and enhancing processing speed. Also, our proposed learning scheme possesses self-correction capabilities that effectively counteract noise during the weight update process, which makes it a promising alternative to traditional error mitigation schemes. We test our framework on two classic navigation tasks - the T-maze and the Morris water-maze - using analogue memristors based on the valence change memory (VCM) effect. Our approach represents a first step towards fully in-memory, online, and error-resilient neuromorphic computing engines based on bio-inspired learning schemes.

show abstract

Continuous learning of spiking networks trained with local rules

Cited by 5 publications

References 47 publications

Efficient Spiking Neural Networks with Sparse Selective Activation for Continual Learning

Efficient Spiking Neural Networks with Sparse Selective Activation for Continual Learning

Event Camera Data Classification Using Spiking Networks with Spike-Timing-Dependent Plasticity

Actor-Critic Networks with Analogue Memristors Mimicking Reward-Based Learning

Contact Info

Product

Resources

About