行为指导的神经种群动力学建模方法论(BLEND)。通过特权知识蒸馏, 在训练时利用行为信号,推理时仅需神经活动。 触发词:神经种群动力学、行为建模、知识蒸馏、特权信息、BLEND、 neural population dynamics, behavior-guided, privileged distillation。
Scanned 9/11/2026
Install to Claude Code
npx -y skills add hiyenwong/ai_collection --skill blend-behavior-guided-neural --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Blend Behavior Guided Neural?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/hiyenwong-blend-behavior-guided-neural)More formats (shields.io, HTML) on the badges page.
---
name: blend-behavior-guided-neural
version: 1.0.0
description: |
行为指导的神经种群动力学建模方法论(BLEND)。通过特权知识蒸馏,
在训练时利用行为信号,推理时仅需神经活动。
触发词:神经种群动力学、行为建模、知识蒸馏、特权信息、BLEND、
neural population dynamics, behavior-guided, privileged distillation。
---
# BLEND: Behavior-guided Neural Population Dynamics
## 核心方法论
### 问题定义
**挑战:** 神经活动与行为联合建模需要复杂设计或过度简化假设。现实场景中,推理时往往只有神经活动数据,没有配对的行为数据。
**解决方案:** BLEND - 行为指导的神经种群动力学建模框架
---
## 关键概念
### 1. 特权知识蒸馏
**核心思想:** 将行为视为特权信息
| 阶段 | 输入 | 模型 |
|------|------|------|
| **训练** | 神经活动 + 行为 | Teacher 模型 |
| **推理** | 仅神经活动 | Student 模型 |
### 2. 框架特点
| 特点 | 说明 |
|------|------|
| **模型无关** | 可增强现有架构,无需从头设计 |
| **无强假设** | 不预设神经-行为的特定关系 |
| **灵活蒸馏** | 多种蒸馏策略可选 |
---
## 性能提升
| 任务 | 提升幅度 |
|------|----------|
| 行为解码 | >50% |
| 神经元身份预测 | >15% |
---
## 应用场景
| 场景 | 说明 |
|------|------|
| **行为解码** | 从神经活动预测行为 |
| **神经元身份预测** | 转录组神经元类型分类 |
| **神经假体** | 实时神经解码 |
| **脑机接口** | 意图解码 |
---
## 技术实现
### PyTorch 示例
```python
import torch
import torch.nn as nn
class BLENDTeacher(nn.Module):
def __init__(self, neural_dim, behavior_dim, hidden_dim):
super().__init__()
self.neural_encoder = nn.Sequential(
nn.Linear(neural_dim, hidden_dim),
nn.ReLU(),
nn.Linear(hidden_dim, hidden_dim)
)
self.behavior_encoder = nn.Sequential(
nn.Linear(behavior_dim, hidden_dim),
nn.ReLU(),
nn.Linear(hidden_dim, hidden_dim)
)
self.fusion = nn.Linear(hidden_dim * 2, hidden_dim)
def forward(self, neural, behavior):
n_feat = self.neural_encoder(neural)
b_feat = self.behavior_encoder(behavior)
return self.fusion(torch.cat([n_feat, b_feat], dim=-1))
class BLENDStudent(nn.Module):
def __init__(self, neural_dim, hidden_dim):
super().__init__()
self.encoder = nn.Sequential(
nn.Linear(neural_dim, hidden_dim),
nn.ReLU(),
nn.Linear(hidden_dim, hidden_dim)
)
def forward(self, neural):
return self.encoder(neural)
```
---
## 与传统方法对比
| 方法 | 推理输入 | 模型设计 | 假设 |
|------|----------|----------|------|
| 联合建模 | 神经+行为 | 复杂 | 强 |
| 单独建模 | 仅神经 | 简单 | 弱 |
| **BLEND** | 仅神经 | 模型无关 | 无 |
---
## 相关技能
- `neuron-model-reconstruction` - 神经元模型重构
- `time-varying-brain-connectivity` - 时变脑网络分析
---
## 来源
- **论文:** BLEND: Behavior-guided Neural Population Dynamics Modeling via Privileged Knowledge Distillation
- **arXiv:** 2410.13872
- **会议:** ICLR 2025
- **效用评分:** 1.0
- **学习日期:** 2026-03-21
## Activation Keywords
- 脑网络分析
- 神经科学方法
- 计算神经科学
- 脑连接建模
## Tools Used
- **read**: Read skill documentation and references
- **exec**: Run analysis scripts and data processing
- **web_fetch**: Fetch papers and resources
## Instructions for Agents
1. Read the skill documentation carefully
2. Understand the methodology and key concepts
3. Apply the techniques to the specific problem
4. Document results and insights
## Examples
```python
# Example usage of the skill methodology
# Refer to the Technical Implementation section for details
```
Is 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!