# 深度学习瓶颈的两条出路：主动推理与离散程序搜索

- 来源：François Chollet (@fchollet)
- 发布时间：2026-08-02 18:30
- AIHOT 分数：45
- AIHOT 链接：https://aihot.virxact.com/items/cmsbnttnh0l6srohvz5ev7wbr
- 原文链接：https://x.com/fchollet/status/2083863030589231160

## AI 摘要

François Chollet 指出，为应对深度学习局限，AI 领域先采用了补丁方案（1）——主动推理，该方案于 2024 年 12 月首次演示，现已无处不在。但他认为长期来看，AI 必然转向方案（2）：用 MDL 原则等更稳健的机制取代 SGD，即彻底告别深度学习（曲线拟合），转向离散程序搜索。

## 正文

To address the limits of deep learning and avoid stalling， the field of AI started by applying patch （1）， which started being demoed 9 months later in December 2024 and has now become completely ubiquitous. However， long term， it is simply inevitable that AI will move to patch （2）.

### 引用推文

> François Chollet：There are essentially two main options to remedy this: 1. Find ways to perform active inference, so that the model adapts its learned program in contact with a ...