推荐系统-总结01:推荐算法发展里程碑

推荐算法从传统方法到深度学习方法发展里程碑

时间 作者 技术名称 说明
2003 Amazon ItemCF(基于物品的协同过滤) 不仅让Amazon的推荐系统广为人知,更让协同过滤成为今后很长时间的研究热点和业界主流的推荐模型。
2006 Netflix MF(矩阵分解) 在Netflix Prize Challenge中,以矩阵分解为主的推荐算法大放异彩,拉开了矩阵分解在业界流行的序幕。
LR(逻辑回归) 深度学习的基础性结构,能够融合多特征,成为了独立于协同过滤的推荐模型的另一个重要发展方向
2010 Rendle FM FM在2012~2014年前后,成为业界主流的推荐模型之一
2015 Yuchi Juan FFM 基于FM提出的FFM在多项CRT预估大赛中夺魁,并被Criteo、美团等公司深度应用在推荐系统、CTR预估等领域
2014 Facebook GBDT+LR 利用GBDT自动进行特征筛选和组合,解决了工程上应用LR的特征组合难度。
2017 阿里巴巴 LS-PLM 自动2012年起成为阿里巴巴主流推荐模型,直至2017年才公开,与三层神经网络及其相似,可以把它看成连接前后深度学习时代的节点。
2015 澳大利亚国立大学 AutoRec 将自编码器的思想与协同过滤结合,提出了一种单隐层神经网络推荐模型
2016 微软 Deep Crossing 深度学习架构在推荐系统中的完整应用
2016 谷歌 Wide&Deep 自提出至今一直在业界发挥着巨大影响力的模型,Wide部分具有逻辑回归的优点,Deep部分具有深度神经网络的优点。
2018 阿里巴巴 DIN 是一次基于实际业务观察的模型改进,提了业界非常知名的深度兴趣网络,把注意力机制引入深度学习推荐模型,用于捕捉用户兴趣。
2019 阿里巴巴 DIEN 兴趣进化网络,DIN的演化版本,引入序列模型,用于模拟用户兴趣进化过程。

本文永久更新地址: https://notlate.cn/p/9e23100d848390f4/

评论

  1. Edge
    1 天前
    2024-12-12 10:42:43

    Your article helped me a lot, is there any more related content? Thanks! https://www.binance.com/en-IN/register?ref=UM6SMJM3

  2. 1 周前
    2024-12-04 6:48:05

    Thanks for sharing. I read many of your blog posts, cool, your blog is very good.

  3. Edge
    2 周前
    2024-12-01 9:38:04

    Thanks for sharing. I read many of your blog posts, cool, your blog is very good.

  4. 2 周前
    2024-11-27 2:53:05

    I don’t think the title of your article matches the content lol. Just kidding, mainly because I had some doubts after reading the article.

  5. 4 周前
    2024-11-18 9:08:09

    Your article helped me a lot, is there any more related content? Thanks!

  6. Edge
    3 月前
    2024-9-12 7:23:30

    Can you be more specific about the content of your article? After reading it, I still have some doubts. Hope you can help me.

  7. 7 月前
    2024-5-07 15:01:56

    Your point of view caught my eye and was very interesting. Thanks. I have a question for you.

  8. 9 月前
    2024-3-28 17:14:22

    I don’t think the title of your article matches the content lol. Just kidding, mainly because I had some doubts after reading the article.

  9. Edge
    9 月前
    2024-3-25 2:29:03

    Thank you for your sharing. I am worried that I lack creative ideas. It is your article that makes me full of hope. Thank you. But, I have a question, can you help me?

  10. Edge
    9 月前
    2024-3-11 14:28:04

    I don’t think the title of your article matches the content lol. Just kidding, mainly because I had some doubts after reading the article.

发送评论 编辑评论


				
|´・ω・)ノ
ヾ(≧∇≦*)ゝ
(☆ω☆)
(╯‵□′)╯︵┴─┴
 ̄﹃ ̄
(/ω\)
∠( ᐛ 」∠)_
(๑•̀ㅁ•́ฅ)
→_→
୧(๑•̀⌄•́๑)૭
٩(ˊᗜˋ*)و
(ノ°ο°)ノ
(´இ皿இ`)
⌇●﹏●⌇
(ฅ´ω`ฅ)
(╯°A°)╯︵○○○
φ( ̄∇ ̄o)
ヾ(´・ ・`。)ノ"
( ง ᵒ̌皿ᵒ̌)ง⁼³₌₃
(ó﹏ò。)
Σ(っ °Д °;)っ
( ,,´・ω・)ノ"(´っω・`。)
╮(╯▽╰)╭
o(*////▽////*)q
>﹏<
( ๑´•ω•) "(ㆆᴗㆆ)
😂
😀
😅
😊
🙂
🙃
😌
😍
😘
😜
😝
😏
😒
🙄
😳
😡
😔
😫
😱
😭
💩
👻
🙌
🖕
👍
👫
👬
👭
🌚
🌝
🙈
💊
😶
🙏
🍦
🍉
😣
Source: github.com/k4yt3x/flowerhd
颜文字
Emoji
小恐龙
花!
上一篇
下一篇