Deep Learning Compact Notes

简介: 一份深度学习的笔记

Linear Regression

Logistic Regression For Classification

Find a linear hyperplane to separate the data, better that output the probability of class

  • Linear model:

$$ z(\boldsymbol{w},\boldsymbol{x}) = \boldsymbol{w} \cdot \boldsymbol{x} $$

  • Link Function:

$$ \hat{p}(z) = \frac{1}{1 + exp(-z)} $$

  • Cross entropy loss:

$$ l(y,\hat{p}) = ylog\hat{p} + ( 1 - y )log(1 - \hat{p} ) $$

  • Cost Function:

$$ L(\boldsymbol{w},\{\boldsymbol{x}_i,y_i\}^m_{i=1} )= \sum^m_{i =1} log(1 + exp(\boldsymbol{w} \cdot \boldsymbol{x}_i)) - y_i\boldsymbol{w} \cdot \boldsymbol{x}_i $$

  • Gradient:

$$ \nabla_w L(\boldsymbol{w},\{\boldsymbol{x}_i,y_i\}^m_{i=1} ) = (\frac{1}{1 + exp(-\boldsymbol{w} \cdot \boldsymbol{x}_i)}- y_i)\boldsymbol{x}_i $$

The backpropagation algorithm works through the layers of deeper neural networks to calculate error gradients w.r.t to weights

目录
相关文章
|
机器学习/深度学习 编解码 数据可视化
Speech Emotion Recognition With Local-Global aware Deep Representation Learning论文解读
语音情感识别(SER)通过从语音信号中推断人的情绪和情感状态,在改善人与机器之间的交互方面发挥着至关重要的作用。尽管最近的工作主要集中于从手工制作的特征中挖掘时空信息,但我们探索如何从动态时间尺度中建模语音情绪的时间模式。
134 0
PointNet++:Deep Hierarchical Feature Learning on Points Sets in a Metrci Space 学习笔记
PointNet++:Deep Hierarchical Feature Learning on Points Sets in a Metrci Space 学习笔记
79 0
|
机器学习/深度学习 传感器 数据挖掘
Review on the Recent Welding Research with Application of CNN-Based Deep Learning
Guo等人16)将CNN应用于线管制造过程中的电阻焊,提出了一种正常焊缝与缺陷焊缝的分类模型,准确率达到99.01%。
106 0
《Investigation of Transformer based Spelling Correction Model for CTC-based End-to-End Mandarin Speech Recognition》电子版地址
Investigation of Transformer based Spelling Correction Model for CTC-based End-to-End Mandarin Speech Recognition
94 0
《Investigation of Transformer based Spelling Correction Model for CTC-based End-to-End Mandarin Speech Recognition》电子版地址
|
机器学习/深度学习 算法 数据挖掘
Paper:He参数初始化之《Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet C》的翻译与解读
Paper:He参数初始化之《Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification》的翻译与解读
|
机器学习/深度学习 自然语言处理 资源调度
【Papernotes】Applying Deep Learning To Airbnb Search
这篇论文详细地记录了 Airbnb 将深度学习引入搜索排名的实践经验,有失败以及对失败的反思,还有在反思的基础上取得的成功。
426 0
【Papernotes】Applying Deep Learning To Airbnb Search
|
决策智能
论文笔记之:Collaborative Deep Reinforcement Learning for Joint Object Search
Collaborative Deep Reinforcement Learning for Joint Object Search   CVPR 2017 Motivation:   传统的 bottom-up object region proposals 的方法,由于提取了较多的 proposal,导致后续计算必须依赖于抢的计算能力,如 GPU 等。
|
机器学习/深度学习 人工智能 自然语言处理
18 Issues in Current Deep Reinforcement Learning from ZhiHu
深度强化学习的18个关键问题   from: https://zhuanlan.zhihu.com/p/32153603     85 人赞了该文章 深度强化学习的问题在哪里?未来怎么走?哪些方面可以突破? 这两天我阅读了两篇篇猛文A Brief Survey of Deep Reinforcement Learning 和 Deep Reinforcement Learning: An Overview ,作者排山倒海的引用了200多篇文献,阐述强化学习未来的方向。
|
机器学习/深度学习 数据挖掘
论文笔记之:Heterogeneous Face Attribute Estimation: A Deep Multi-Task Learning Approach
Heterogeneous Face Attribute Estimation: A Deep Multi-Task Learning Approach  2017.11.28    Introduction:   人脸属性的识别在社会交互,提供了非常广泛的信息,包括:the person’s ...
Deep Learning vs. Machine Learning vs. Pattern Recognition
Deep learning, machine learning, and pattern recognition are highly relevant topics commonly used in the field of robotics with artificial intelligence.
4542 0