48 lines
1.8 KiB
Plaintext
48 lines
1.8 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "f0f1791b",
|
|
"metadata": {
|
|
"origin_pos": 0
|
|
},
|
|
"source": [
|
|
"# 多层感知机\n",
|
|
":label:`chap_perceptrons`\n",
|
|
"\n",
|
|
"在本章中，我们将第一次介绍真正的*深度*网络。\n",
|
|
"最简单的深度网络称为*多层感知机*。多层感知机由多层神经元组成，\n",
|
|
"每一层与它的上一层相连，从中接收输入；\n",
|
|
"同时每一层也与它的下一层相连，影响当前层的神经元。\n",
|
|
"当我们训练容量较大的模型时，我们面临着*过拟合*的风险。\n",
|
|
"因此，本章将从基本的概念介绍开始讲起，包括*过拟合*、*欠拟合*和模型选择。\n",
|
|
"为了解决这些问题，本章将介绍*权重衰减*和*暂退法*等正则化技术。\n",
|
|
"我们还将讨论数值稳定性和参数初始化相关的问题，\n",
|
|
"这些问题是成功训练深度网络的关键。\n",
|
|
"在本章的最后，我们将把所介绍的内容应用到一个真实的案例：房价预测。\n",
|
|
"关于模型计算性能、可伸缩性和效率相关的问题，我们将放在后面的章节中讨论。\n",
|
|
"\n",
|
|
":begin_tab:toc\n",
|
|
" - [mlp](mlp.ipynb)\n",
|
|
" - [mlp-scratch](mlp-scratch.ipynb)\n",
|
|
" - [mlp-concise](mlp-concise.ipynb)\n",
|
|
" - [underfit-overfit](underfit-overfit.ipynb)\n",
|
|
" - [weight-decay](weight-decay.ipynb)\n",
|
|
" - [dropout](dropout.ipynb)\n",
|
|
" - [backprop](backprop.ipynb)\n",
|
|
" - [numerical-stability-and-init](numerical-stability-and-init.ipynb)\n",
|
|
" - [environment](environment.ipynb)\n",
|
|
" - [kaggle-house-price](kaggle-house-price.ipynb)\n",
|
|
":end_tab:\n"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"language_info": {
|
|
"name": "python"
|
|
},
|
|
"required_libs": []
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 5
|
|
} |