Yahav Bechavod

Postdoc @Penn | Fmr @Apple Scholar in AI/ML | PhD @HebrewU. Interests in ML, algorithmic game theory, and fairness.

[Yahav Bechavod @yahavbe]

Today (𝟐𝐩𝐦-𝟑:𝟑𝟎𝐩𝐦 𝐇𝐒𝐓 #𝟕𝟏𝟐) at #ICML2023 --- presenting our new work (joint with @Aaroth).

We give algorithms for learning with individual fairness even when label feedback is restricted to positive predictions only.

Come say hi!

[Elad Granot @eladgranot]

Did you think your model is protected if you publish it as a black box? Think twice.

Happy to share our #ICLR2023 paper:
An Exact Poly-Time Membership-Queries Algorithm for Extraction a three-Layer ReLU Network
https://t.co/urlchma0vM

w. Amit Daniely

[Yahav Bechavod @yahavbe]

Today 9PST / 12EST at TOC4Fairness! Come say hi!

[toc4fairness @toc4fairness]

Hi everyone, we are restarting the TOC4Fairness seminar series this Wednesday, February 15th (9-10am pst, 12-1pm est)! We are excited to have @yahavbe
as our first speaker of the semester.

More info here: https://t.co/O57VYYDpAB

[Aaron Roth @Aaroth]

A new paper! Multicalibration is an exciting recent solution concept from the algorithmic fairness literature. Boosting is a classical learning technique. It turns out there is a tight connection between multicalibration and boosting for squared error regression. A thread:🧵

[European Workshop on Algorithmic Fairness @EWAFWorkshop]

🚨 Call for Papers 🚨 Excited to get things started for #EWAF2023 We're looking for interdisciplinary research focused on algorithmic fairness in the European context. Join us from June 7 to 9 in Switzerland 🇨🇭 Submission deadline on Feb. 9 🤖

https://t.co/OOZmT81J2j

[Yahav Bechavod @yahavbe]

Thank you @InbalTalgam and Irene Lo for the invitation!

[Yahav Bechavod @yahavbe]

I recently conducted an interview with @JubaZiani for the newest edition of @AcmSIGecom #SIGecomExchanges. We talked about algorithmic fairness, the effects of strategic behavior, feedback loops, and wealth accumulation over generations. Find it here: https://t.co/r06XbaHKzR

[Gautam Kamath @thegautamkamath]

Aaron Roth (@aaroth) just wrapped up a brand new course on uncertainty estimation in machine learning. Looks great!

Featuring 160 pages of "notes" (pretty much a book): https://t.co/CEqnpae8Gk
Full lecture videos: https://t.co/3iSFSTbBgy
Catchy website: https://t.co/wEsn5wakF8.

[Yahav Bechavod @yahavbe]

Happening today 12pm ET at the @toc4fairness seminar! Come hear @charapod talk about our recent work "Information Discrepancy in Strategic Learning" w/ @zstevenwu and @JubaZiani!

https://t.co/zExI2oJGus

[Chara Podimata @charapod]

Tomorrow at 9am PST / 12pm EST at TOC4Fairness!

I'm giving a talk on "Information Discrepancy in Strategic Learning", my joint work with @yahavbe, @zstevenwu, and @JubaZiani that appeared in ICML22. Come say hi!

[Yahav Bechavod @yahavbe]

At #icml22. Feels great to finally be back to an in-person conference edition! Talking about this paper tomorrow (Thursday) 11:40 at room 307, and presenting the poster tomorrow 6pm-8pm at Hall E #1218. Come say hi!

[Yahav Bechavod @yahavbe]

Excited about this paper with @charapod, @zstevenwu, and Juba Ziani. We study learning with strategic behavior, but with a twist – the decision rule is not exposed to strategic individuals, who instead attempt to learn about it from their peers. https://t.co/iTOI6pKTYE

[Yahav Bechavod @yahavbe]

Accepted at #ICML2022 :)

[Yahav Bechavod @yahavbe]

Excited about this paper with @charapod, @zstevenwu, and Juba Ziani. We study learning with strategic behavior, but with a twist – the decision rule is not exposed to strategic individuals, who instead attempt to learn about it from their peers. https://t.co/iTOI6pKTYE

[Yahav Bechavod @yahavbe]

Excited about this paper with @charapod, @zstevenwu, and Juba Ziani. We study learning with strategic behavior, but with a twist – the decision rule is not exposed to strategic individuals, who instead attempt to learn about it from their peers. https://t.co/iTOI6pKTYE

[Yahav Bechavod @yahavbe]

The videos from our NeurIPS'21 #StratML workshop are now available! Watch the invited talks and panel discussions here:
https://t.co/FRflAK0QPu

[Yahav Bechavod @yahavbe]

The list of #StratML invited talks is out!
https://t.co/3JIjSlgHtD

[Aaron Roth @Aaroth]

Mark your calendars (February 23) for the SIGEcom winter meeting: a free virtual day of tutorials, talks, and conversations on the intersection of algorithmic fairness and game theory/economics. We've put together a stellar lineup. I personally can't wait! https://t.co/OQQg6rVN6W

[Yahav Bechavod @yahavbe]

The list of #StratML invited talks is out!
https://t.co/3JIjSlgHtD

[Yahav Bechavod @yahavbe]

Very glad to announce our workshop, "Learning and Decision-Making with Strategic Feedback (StratML'21)" at #NeurIPS2021! We have a wonderful collection of invited speakers. The full schedule and CFP (deadline: Sep. 17) are here: https://t.co/PezIq61LB6

[Yahav Bechavod @yahavbe]

The list of accepted papers to #StratML is now available! See the contributed talks and posters here:
https://t.co/8uRxjrfB1N

[Yahav Bechavod @yahavbe]

Very glad to announce our workshop, "Learning and Decision-Making with Strategic Feedback (StratML'21)" at #NeurIPS2021! We have a wonderful collection of invited speakers. The full schedule and CFP (deadline: Sep. 17) are here: https://t.co/PezIq61LB6

[Jonathan Ullman @thejonullman]

I gave a talk @SimonsInstitute about the complexity of differential privacy and you can watch it here: https://t.co/t2YviLU87t

I tried something new by giving a survey of the major questions, rather than a talk about a recent paper. I, for one, thought it went well.

[Yahav Bechavod @yahavbe]

With my amazing co-organizers: @CeleMenDu, Eric Mazumdar, @HodaHeidari, and Tijana Zrnic.

[Yahav Bechavod @yahavbe]

Very glad to announce our workshop, "Learning and Decision-Making with Strategic Feedback (StratML'21)" at #NeurIPS2021! We have a wonderful collection of invited speakers. The full schedule and CFP (deadline: Sep. 17) are here: https://t.co/PezIq61LB6

[Yahav Bechavod @yahavbe]

New entries on the TOC4Fairness blog:

"Fair Clustering with Probabilistic Group Membership", by Seyed A. Esmaeili.

"Self-fulfilling and self-negating predictions: a short tale of performativity in machine learning", by Tijana Zrnic.

https://t.co/xUS3NC4nUq

Check these out!!