Abstract:Spike Timing-Dependent Plasticity (STDP) is a promising substitute to backpropagation for local training of Spiking Neural Networks (SNNs) on neuromorphic hardware. STDP allows SNNs to address classification tasks by combining unsupervised STDP for feature extraction and supervised STDP for classification. Unsupervised STDP is usually employed with Winner-Takes-All (WTA) competition to learn distinct patterns. However, WTA for supervised STDP classification faces unbalanced competition challenges. In this paper, we propose a method to effectively implement WTA competition in a spiking classification layer employing first-spike coding and supervised STDP training. We introduce the Neuronal Competition Group (NCG), an architecture that improves classification capabilities by promoting the learning of various patterns per class. An NCG is a group of neurons mapped to a specific class, implementing intra-class WTA and a novel competition regulation mechanism based on two-compartment thresholds. We incorporate our proposed architecture into spiking classification layers trained with state-of-the-art supervised STDP rules. On top of two different unsupervised feature extractors, we obtain significant accuracy improvements on image recognition datasets such as CIFAR-10 and CIFAR-100. We show that our competition regulation mechanism is crucial for ensuring balanced competition and improved class separation.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: How to effectively implement the WTA (Winner - Takes - All) competition mechanism through supervised STDP (Spike Timing - Dependent Plasticity) in spike - based classification tasks. Specifically, the author proposes a new architecture - Neuronal Competition Group (NCG) - to improve the classification performance in Spiking Neural Networks (SNNs) trained with first - spike coding and supervised STDP. ### Main problems: 1. **Unbalanced competition challenge**: The traditional WTA mechanism faces the problem of unbalanced competition in supervised STDP classification, which causes some neurons to dominate weight updates while other neurons have difficulty learning effective patterns. 2. **Learning of intra - class diversity**: In order to improve the classification ability, it is necessary to promote neurons within each class to learn different patterns instead of all neurons learning the same pattern. 3. **Fair decision - making mechanism**: In the decision - making process, different neurons may use different thresholds, which may lead to unfair decisions. ### Solutions: The author introduces the **Neuronal Competition Group (NCG)** architecture, whose main features include: - **Intra - class WTA mechanism**: Each class is mapped to an NCG, and intra - class WTA competition is implemented among neurons within the NCG, ensuring that only one neuron updates its weight each time. - **Competition regulation mechanism**: A competition regulation mechanism based on a two - compartment threshold is introduced to ensure the balance of competition among intra - class neurons and fairness during decision - making. The specific formula is as follows: \[ \Delta \theta'_j = \begin{cases} +\eta_{th} \cdot \frac{M - 1}{M} & \text{if } t_j = \min\{t_1, \dots, t_M\} \\ -\eta_{th} \cdot \frac{1}{M} & \text{otherwise} \end{cases} \] where $\theta'_j$ is the training threshold of neuron $n_j$, $\theta_j$ is the testing threshold, $M$ is the number of neurons in the NCG, $\eta_{th}$ is the threshold learning rate, and $t_j$ is the spike timestamp of neuron $n_j$. - **Fixed test threshold**: All neurons use a fixed test threshold $\theta$ during inference to ensure fair decision - making. Through these improvements, the author has successfully solved the unbalanced competition problem in supervised STDP classification and significantly improved the classification accuracy on image recognition datasets such as CIFAR - 10 and CIFAR - 100. ### Experimental results: Experiments show that the NCG architecture has achieved significant performance improvements on multiple datasets, especially on complex datasets such as CIFAR - 10 and CIFAR - 100. For example, on the CIFAR - 100 dataset, the accuracy of the S2 - STDP+NCG method has increased by 6.51 percentage points. In conclusion, this paper effectively solves the unbalanced competition problem in supervised STDP classification and significantly improves the performance of spike - based classification tasks by introducing the NCG architecture.

Neuronal Competition Groups with Supervised STDP for Spike-Based Classification

Paired Competing Neurons Improving STDP Supervised Local Learning In Spiking Neural Networks

Toward Efficient Processing and Learning with Spikes: New Approaches for Multispike Learning

SSTDP: Supervised Spike Timing Dependent Plasticity for Efficient Spiking Neural Network Training

Unsupervised Learning Based on Temporal Coding Using STDP in Spiking Neural Networks.

An unsupervised STDP-based spiking neural network inspired by biologically plausible learning rules and connections

Unsupervised Learning of Spike-Timing-Dependent Plasticity Based on a Neuromorphic Implementation

An STDP-Based Supervised Learning Algorithm for Spiking Neural Networks

RRAM-Based Spiking Neural Network with Target-Modulated Spike-Timing-Dependent Plasticity

Deep Unsupervised Learning Using Spike-Timing-Dependent Plasticity

Supervised Learning with First-to-Spike Decoding in Multilayer Spiking Neural Networks

BP-STDP: Approximating backpropagation using spike timing dependent plasticity

Fast and Accurate Classification with a Multi-Spike Learning Algorithm for Spiking Neurons.

Pattern Recognition Computation in a Spiking Neural Network with Temporal Encoding and Learning

Temporal Competitive Learning Induced in Neural Networks by Spike Timing-Dependent Plasticity

Training Deep Spiking Convolutional Neural Networks With STDP-Based Unsupervised Pre-training Followed by Supervised Fine-Tuning

Efficient visual object representation using a biologically plausible spike-latency code and winner-take-all inhibition

Desire Backpropagation: A Lightweight Training Algorithm for Multi-Layer Spiking Neural Networks based on Spike-Timing-Dependent Plasticity

Fast Unsupervised Learning for Visual Pattern Recognition Using Spike Timing Dependent Plasticity.

Spatiotemporal Backpropagation based on Channel Reward for Training High-Precision Spiking Neural Network

S-TLLR: STDP-inspired Temporal Local Learning Rule for Spiking Neural Networks