Open in app

Sign in

Write

Sign in

Arthur Lee
Arthur Lee

25 Followers

Home

About

Apr 14

2021 千辛萬苦 申請綠卡過程

拿到綠卡 是許多人在美國的里程碑之一 我也很幸運在2020抽到H1B 可惜我是CP 所以我只好回台灣面美簽拿H1B了 才能開始跑綠卡流程 此篇文章會分享 綠卡申請timeline 注意事項 2021/1/13 我回到美國,H1B開始activate。 2021/1/14 公司律師開始邀請我填問卷,上傳資料,我立刻填完,並且上傳資料 2021/1/15 我寄信給律師要求fo …

Eb2綠卡

4 min read

2021 千辛萬苦 申請綠卡過程
2021 千辛萬苦 申請綠卡過程
Eb2綠卡

4 min read


Jan 15

[Eng blog: Pinterest-3] How Pinterest Leverages Realtime User Actions in Recommendation to Boost Homefeed Engagement Volume

🤗 Machine learning Engineer blog challenge (3/100) Pinterest — 2022 How Pinterest Leverages Realtime User Actions in Recommendation to Boost Homefeed Engagement Volume | by Pinterest Engineering | Pinterest Engineering Blog | Medium 🤔 What problem do they solve? 他們想要找到一個single去抓出short-term user interest並且把它納入目前的Pins HomeFeed ranking model裡面,有了short-term user interest搭配ML application可以實現responsive的功能(app可以快速反應使用者的短期篇好,例如點了日本Feed,過幾秒刷新Feed,內容會多了日本Feed)

Pinterest

3 min read

[Eng blog: Pinterest-3] How Pinterest Leverages Realtime User Actions in Recommendation to Boost…
[Eng blog: Pinterest-3] How Pinterest Leverages Realtime User Actions in Recommendation to Boost…
Pinterest

3 min read


Nov 5, 2022

United Airline last name多了SR問題

摘要 這篇主要會談到,如何被UA系統坑了,ticket上名字多了”SR”,還有韓亞航空服務品質的經驗,還會介紹下次Action Item,如何避免 過程 幾個月前就定好了來回機票, 用UA點數訂的 回來的時候搭韓亞航空,發現在UA上面不能選位子, (ANA是可以),所以特別去韓亞航空官方網站用我的confirmation number去查,發現一直找不到,想說可能系統更新有delay,拖了一個月 今天又再次挑戰,直接打電話給韓亞航空,電話接上了,服務態度很差,過程中發現我的名字last name多了”SR”,所以不給過,我記得當時在UA訂票時候,我一直想改掉但一直不給改,本以為沒差,沒想到有差 後來打給UA,客服說如果是其他航空的話要打去其他航空,如果是UA自己的話,可以幫忙改,很快速(1–3 min)幫我改好了 雖然改好了國內UA線,但韓亞還是沒改,明天早起再打電話去改 Root Cause

United Airlines

2 min read

United Airline last name多了SR問題
United Airline last name多了SR問題
United Airlines

2 min read


Aug 8, 2022

[Eng blog: Doordash-22] Using a Multi-Armed Bandit with Thompson Sampling to Identify Responsive Dashers

🤗 Machine learning Engineer blog challenge (2/100) Doordash — 2022 Using a Multi-Armed Bandit with Thompson Sampling to Identify Responsive Dashers 🤔 What problem do they solve? 他們想要找到一個model去決定要對哪些dasher(運送員)發訊息去送貨,特別在尖峰時刻,supply 不夠多時候,如下圖所示

Machine Learning

2 min read

[Eng blog: Doordash-22] Using a Multi-Armed Bandit with Thompson Sampling to Identify Responsive…
[Eng blog: Doordash-22] Using a Multi-Armed Bandit with Thompson Sampling to Identify Responsive…
Machine Learning

2 min read


Jul 25, 2022

西南航空apply travel Fund issue [solved]-2022

這篇是最近遇到的一個小問題 我想要apply travel fund,但一直出現錯誤 如果其他人也有遇到,希望也可以幫助到大家 Solution 假設你的名字如下 First Name: Young Xun Last Name: Lee 記得改兩個地方! Who’s flying (我一開始這沒改,系統一直抱錯) Young / Xun / Lee Flight Credits Young / Lee

西南航空

2 min read

西南航空apply travel Fund issue [solved]-2022
西南航空apply travel Fund issue [solved]-2022
西南航空

2 min read


Jul 23, 2022

[Eng blog: Doordash-21] Using Triplet Loss and Siamese Neural Networks to Train Catalog Item Embeddings

🤗 Machine learning Engineer blog challenge (1/100) Doordash — 2021 Using Triplet Loss and Siamese Neural Networks to Train Catalog Item Embeddings (doordash.engineering) 🤔 What problem do they solve? 他們想要建構出 catalog(品項) embedding 首先什麼是catalog in Doordash (可能有誤解)? 同一家餐廳有很多的品項, 像是salmon sushi, boba tea, 而不同餐廳也可以share同個品項,每一家珍奶店都有個東西叫 “boba tea” 讓boba tea 有自己的representation, 不同家的boba tea share 一樣的representation -> 裡面blog可能沒明說(也許我看漏) 但我從他們後來用品項去建構store embedding,應該是考慮這狀況,而且這樣更可以solve cold-start issue (new store)

Machine Learning

3 min read

[Eng blog: Doordash-21] Using Triplet Loss and Siamese Neural Networks to Train Catalog Item…
[Eng blog: Doordash-21] Using Triplet Loss and Siamese Neural Networks to Train Catalog Item…
Machine Learning

3 min read


Jul 10, 2022

KDD 21':Understanding and Improving Fairness-Accuracy Trade-offs in Multi-Task Learning (google)

🤗 Recommendation system paper challenge (31/50) paper link Google KDD 2021 🤔 What problem do they solve? 現有的MTL (multi-task learning) 沒有針對fairness進行優化,只有STL (single task learning)針對fairness進行優化 所以這篇paper在於解決 fairness on MTL的問題 😎 Contribution 定義新的metric: ARFG: 對每個task去計算MTL中的FPRGap跟STL的FPRGap的比值,在平均起來 ARE: 對每個task去計算MTL中的error rate跟STL的error rate的比值,在平均起來

Kdd2021

4 min read

KDD 21':Understanding and Improving Fairness-Accuracy Trade-offs in Multi-Task Learning (google)
KDD 21':Understanding and Improving Fairness-Accuracy Trade-offs in Multi-Task Learning (google)
Kdd2021

4 min read


Jul 7, 2022

KDD 21':Learning to Embed Categorical Features without Embedding Tables for Recommendation (google)

🤗 Recommendation system paper challenge (30/50) paper link Google KDD 2021

Google

10 min read

KDD 21':Learning to Embed Categorical Features without Embedding Tables for Recommendation (
KDD 21':Learning to Embed Categorical Features without Embedding Tables for Recommendation (
Google

10 min read


May 14, 2022

Hilton Aspire 航空報銷 手把手

Hilton Aspire link: AmEx Hilton Aspire 信用卡【2021.11 更新:Targeted 消费 8k 送一张 FN;150k 开卡奖励】 — 美国信用卡指南 (uscreditcardguide.com) 最近我申請了這張卡 算是我第一個高階卡吧 之前從來沒有報銷過航空 傻傻分不清 怎麼報銷 如果小夥伴有同樣的問題 也許可以參考這一篇 Step1: 註冊要報銷的航空 首先先去Amex 官方網站 Log In to My Account | American Express US 點選 “Benefits” 點選”Travel”, 找到 “$250 Airline Fee Credit”

信用卡

3 min read

Hilton Aspire 航空報銷 手把手
Hilton Aspire 航空報銷 手把手
信用卡

3 min read


Jan 3, 2022

KDD 19': Sampling-bias-corrected neural modeling for large corpus item recommendations

🤗 Recommendation system paper challenge (29/50) paper link 🤔 What problem do they solve? Large Scale item recommendation Given a query x, we would like to recommend items y from M items and we can observe reward (watch time) for each pair (x, y) Label: reward r (watch time) Train data: a pair (query x, item y, reward r) Model: two-tower model…

Kdd 2019

3 min read

KDD 21': Sampling-bias-corrected neural modeling for large corpus item recommendations
KDD 21': Sampling-bias-corrected neural modeling for large corpus item recommendations
Kdd 2019

3 min read

Arthur Lee

Arthur Lee

25 Followers

An machine learning engineer in Bay Area in the United States

Following
  • Nana Chiang

    Nana Chiang

  • 數據分析那些事

    數據分析那些事

  • Pinterest Engineering

    Pinterest Engineering

  • Skyscanner Engineering

    Skyscanner Engineering

  • Tim O'Reilly

    Tim O'Reilly

See all (34)

Help

Status

About

Careers

Blog

Privacy

Terms

Text to speech

Teams