- shima__shima
- 3578
- 12
- 0
- 0
Great minds for a great topic #FAT* #fairness @fatconference pic.twitter.com/lrUwFEblSp
2020-01-27 23:07:00Fascinating discussion of survey results in "Assessing the intersection of Organizational Structure and FAT* efforts within industry" fatconference.org/2020/acceptedt… #FAT2020 pic.twitter.com/iyFIdlTRuX
2020-01-27 23:40:20A bit shaken after a 16 hour flight and heartbreaking news #RIPKobe but holding it together at @fatconference in Barcelona. ML Ethics folks, who else is around? 👋🏻 pic.twitter.com/GWfxOB4YhY
2020-01-27 23:43:10"What if @propublica found out?" @borirakova and @jingyingyang explain the challenges to organizational change within private tech companies in changing practices and approaches to "ethical" and "fair" machine learning and AI #FAT2020 @fatconference pic.twitter.com/zLijqX4veb
2020-01-28 00:26:46Ready to start our #FAT2020 tutorial: can an algorithmic system be a friend to a police officer's discretion? pic.twitter.com/MQX91yZ97A
2020-01-28 00:43:01Tomorrow morning at @fatconference in the 9-10:50 accountability session @MikeKatell will be presenting the latest work from @critplat! "Toward Situated Interventions for Algorithmic Equity: Lessons from the Field" #FAT2020 (Thread) pic.twitter.com/UOVUy9O7ty
2020-01-28 00:57:51A very interesting and thought-provoking tutorial on gender and the GDPR by @FusterGloria at @fatconference pic.twitter.com/00ASMuKzEM
2020-01-28 01:08:43No references to gender in the GDPR, as @FusterGloria adopts the @tnhh regex legal analysis method. #FAT2020 pic.twitter.com/NlBYoHfjO8
2020-01-28 01:39:53Very insightful session in measurement of bias for NLP systems ( lessons for NLP systems). Thanks @sulin_blodgett and @az_jacobs to share your research with us. #FAT2020 #NLP #AI pic.twitter.com/nGr35QjYgm
2020-01-28 02:53:27Thanks to everyone who made it to the librec-auto tutorial at #FAT2020 today! If you missed it don't worry, you can still try out Fairness-Aware librec on your own 🤓 Link to the tutorial video will be coming soon! docs.google.com/document/d/1yb… pic.twitter.com/LEwfOtyMU4
2020-01-28 03:57:06Reading up on HR+AI+Fairness, w/ #FAT2020 in my TL and my new role at @Randstad nearing 😅 "Realistically, we have a greater ability to ensure both accuracy and fairness in AI systems than we do to influence or enlighten recruiters and hiring managers." hbr.org/2019/11/buildi…
2020-01-28 04:32:30FAT*2020の登録は680らしい.NeurIPS2014併設で,床面積の80%を占める面積が深層学習ワークショップが割り当てられる横で,30〜50人ほどでFATMLワークショップをやってたときと隔世の感 twitter.com/ChaToX/status/…
2020-01-28 05:07:25@fatconference Hello! One of the General Co-Chairs here. We have 680 registered participants. We will be taking questions via Etherpads pad.riseup.net/p/fatqa - people are obviously free to use other chat applications or channels during the conference.
2020-01-27 15:54:04チュートリアル Explainable AI in Industry - Practical Challenges and Lessons Learned sites.google.com/view/fat20-exp…
2020-01-28 05:08:12LinkedIn の中の人によるチュートリアル.解釈可能性チートシート: twitter.com/chandan_singh9… はここで紹介されてた 解釈可能モデルを使って,特徴のどんなところがまずいのかとかを修正する手順とかを紹介していた
2020-01-28 05:08:30Made a scikit-style cheat sheet for interpretability: github.com/csinva/csinva.… It's certainly not exhaustive, but hopefully it's useful! #ML #DataScience #AI #MachineLearning #Statistics pic.twitter.com/uBfgzAYs4F
2019-06-12 23:14:14LIME の改良版 LIMEx を述べてた.情報のないベースラインから元画像などに変化させていったときにLIMEみたいにクラス分類の勾配を調べ,その勾配を積分するという integrated gradients.出力にいつも効いている部分がより明確にでるらしい. proceedings.mlr.press/v70/sundararaj…
2020-01-28 05:08:58チュートリアル From the Total Survey Error Framework to an Error Framework for Digital Traces of Humans tinyurl.com/ted2020
2020-01-28 05:09:13計算社会学で,twitterとかのツイートから,トランプ大統領の支持を調査するとかする場合の注意点について. - validity:信号選択の妥当性 - platform affordance error: プラットフォーム依存の誤差(クエリ拡張,文字数制限) - signal selection error: 観測する信号が目的とあっていない
2020-01-28 05:09:32- platform coverage error:目標母集団と実際の利用者の差 - entity selection error: USの住人を目標とするときに,英語のツイートを収集するとヒスパニックの人が外れる - signal processing error:前処理,フィルタリングの問題 - adjustment error: 標本の重み付けなどの影響
2020-01-28 05:09:45I’m ready for this- @MarankeWieringa first up this morning @fatconference twitter.com/marankewiering… pic.twitter.com/dMz4LZa3ig
2020-01-28 16:45:21What is accountability, asks @fatconference. Join our session at 9.00 to find out. #FAT2020 pic.twitter.com/ZasWheRbuV
2020-01-28 16:33:47Amazing expansion for this @fatconference #FAT2020 pic.twitter.com/23Cn1zFC93
2020-01-28 17:07:43Algorithmic accountability: an algorithm as a socio-technical system rather than only the technical aspect. Actor, Forum, Relationship @MarankeWieringa @fatconference 200+ papers included
2020-01-28 17:19:09"Algorithmic realism" presented by @salome_viljoen_ a legal perspective to understand algorithms. CS and Law tend to focus on "formalisms": principles, axioms, norms to derive reasoning. Realism, instead, looks at the context and infers principles @fatconference
2020-01-28 17:24:34glad you could make it to #FAT2020, @npseaver! (@MarankeWieringa) pic.twitter.com/0tq5EOCjtq
2020-01-28 17:28:54Algorithmic accountability in public administration in Europe by @GDPRgirl at @fatconference GDPR protects against dystopians such as "social credit systems", or does it? Plenty of language permitting automated decision making. Fixes are needed against discrimination and biases
2020-01-28 17:32:07