20260518-morning:新增内容

This commit is contained in:
2026-05-18 10:17:56 +08:00
parent b116710e4c
commit 2faf4bb002
38 changed files with 1528 additions and 2 deletions

View File

@@ -0,0 +1,41 @@
---
title: "Composable Base Model Architecture"
created: 2026-05-15
updated: 2026-05-15
type: concept
tags: [machine-learning, architecture, modular-design]
sources: [raw/papers/zeng-neurida-2025.md]
---
# Composable Base Model Architecture
**可组合基础模型架构**是 NeurIDA 实现 [[dynamic-in-database-modeling|动态建模]] 的架构基础。
## 构成
```
基础模型池 M = {m₁, m₂, ..., mₖ}
+
共享模型组件
├── 统一元组编码器Unified Tuple Encoder
├── 关系感知消息传递模块Relation-Aware Message Passing
└── 上下文感知融合模块Context-Aware Fusion
```
## 基础模型池
涵盖四类异构模型:
- **Traditional ML**RF, CatBoost, LightGBM, Logistic Regression
- **Tuple Representation Models (TRM)**FT-Transformer, ARM-Net, TabM, ResNet, DNN, DeepFM
- **Tabular Foundation Models**TabPFN, TabICL
- **Large Tabular Models (LTM)**TP-BERTa, Nomic, BGE
## 设计原则
1. **异构性**:不同架构的模型互补,适应不同数据分布
2. **可组合性**:共享组件的接口统一,可任意与基础模型组合
3. **查询条件化**[[conditional-model-dispatcher|Dispatcher]] 根据任务选择合适的 m*DIME 按需装配
## 来源
- [[zeng-neurida-2025|NeurIDA 论文]]