梯度突触可塑性稀疏实现方法论。实现真正稀疏且通用的梯度突触可塑性规则,保持在线学习能力。适用于脉冲神经网络、在线学习、突触可塑性研究。触发词:突触可塑性、梯度下降、稀疏实现、在线学习、sparse plasticity、gradient-based、online learning。
Scanned 9/11/2026
Install to Claude Code
npx -y skills add hiyenwong/ai_collection --skill sparse-gradient-plasticity --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Sparse Gradient Plasticity?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/hiyenwong-sparse-gradient-plasticity)More formats (shields.io, HTML) on the badges page.
---
name: sparse-gradient-plasticity
description: 梯度突触可塑性稀疏实现方法论。实现真正稀疏且通用的梯度突触可塑性规则,保持在线学习能力。适用于脉冲神经网络、在线学习、突触可塑性研究。触发词:突触可塑性、梯度下降、稀疏实现、在线学习、sparse plasticity、gradient-based、online learning。
user-invocable: true
---
# Sparse Gradient Synaptic Plasticity
## 核心思想
实现真正稀疏且通用的梯度突触可塑性规则,避免手动推导梯度或牺牲在线能力。
**来源:** arXiv:2501.11407
**效用:** 0.92
---
## 实现
```python
import numpy as np
class SparseGradientPlasticity:
def __init__(self, n_pre, n_post, sparsity=0.1):
self.W = np.random.randn(n_post, n_pre) * 0.1
self.mask = np.random.rand(n_post, n_pre) < sparsity
self.W *= self.mask
self.eta = 0.01
def forward(self, x):
return self.W @ x
def update(self, pre, post, error):
dW = self.eta * np.outer(error * post, pre)
self.W += dW * self.mask
```
---
## Activation Keywords
- 突触可塑性
- 梯度下降
- 稀疏实现
## Tools Used
- numpy
## Instructions for Agents
1. 创建稀疏连接掩码
2. 实现在线梯度更新
3. 保持稀疏性约束
## Examples
在线学习任务中的突触权重更新。
## 参考文献
- arXiv:2501.11407Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!