Ryan (@RowRyRo) profile picture

Ryan

@RowRyRo

privacy research engineer

Joined October 2019
  • Tweets130
  • Following131
  • Followers258
12Media
Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)Photo by Ryan (@RowRyRo)
Another result of Durfee and @RowRyRo that should be better known. They showed that the exponential mechanism in DP has better composition properties than generic DP algorithms. One application: Selecting the top k elements of some set. It turns out you can do it in one shot.
Differential Privacy (@DiffPriv) profile pictureDifferential Privacy@DiffPriv9 Aug 2021
One-shot DP Top-k mechanisms ift.tt/3sglVpD
One-shot DP Top-k mechanisms

One-shot DP Top-k mechanisms

In the last blog post, we showed that the exponential mechanism enjoys improved composition bounds over general pure DP mechanisms due to a property called bounded range. For this post, we will...

differentialprivacy.org
1
9
Originating from conversations at the 2020 Google Workshop on Federated Learning and Analytics (shorturl.at/myAY8), 26 @GoogleAI researchers and 27 academic researchers began a year-long collaboration resulting in A Field Guide to Federated Optimization
Image from Peter Kairouz (@KairouzPeter)
5
44
3
213
An important result that has gone under-appreciated for a few years: the exponential mechanism is about twice as private as you think (at least under composition).
Jonathan Ullman (@thejonullman) profile pictureJonathan Ullman@thejonullman12 Jul 2021
Cool new post by @shortstein and @RowRyRo showing how to save a factor of 2 in the privacy analyss of the exponential mechanism! Fancy new algorithms are great, but improving a ubiquitous tool with no downsides is hard to beat! differentialprivacy.org/expo…
A Better Privacy Analysis of the Exponential Mechanism

A Better Privacy Analysis of the Exponential Mechanism

A basic and frequent task in data analysis is selection – given a set of options \(\mathcal{Y}\), output the (approximately) best one, where “best” is defined by some loss function \(\ell : \mathca...

differentialprivacy.org
2
11
CfP for the workshop on Theory and Practice of Differential Privacy (TPDP 2021, co-located with @icmlconf #ICML2021) is up! Deadline: May 28. Submit work on any aspects of DP! Co-chaired with @radcummings. Invited speakers are Katrina Ligett and @RowRyRo. tpdp.journalprivacyconfident…
Image from Gautam Kamath (@thegautamkamath)
Image from Gautam Kamath (@thegautamkamath)
Image from Gautam Kamath (@thegautamkamath)
Image from Gautam Kamath (@thegautamkamath)
1
13
2
43
Did you miss TPDP 2020? If you were unable to attend, or want to watch again, videos of the talks are available here: tpdp.journalprivacyconfident…
5
1
18
The **Algorithms** group @MSFTResearch Redmond is hiring! It is the sister group to our ML Foundations group, and has already strong presence in differential privacy and coding! Post Doc: careers.microsoft.com/us/en/… FTE: careers.microsoft.com/us/en/…
22
93
📺🎙 Today’s privacy_infra() #PrivacyEng event is about to begin (10am PT)! Join us at zoom.us/webinar/register/801…
Image from Transcend (@transcend_io)
2
1
2
🗓We've got our last #privacyinfra event for 2020 this Thursday at 10am PT, with privacy eng. talks from LinkedIn, Wikipedia, and more. #privacyeng Register here: zoom.us/webinar/register/481…
2
4
A year after 'The Ethical Algorithm: The Science of Socially Aware Algorithm Design' was released, we asked @mkearnsupenn and @Aaroth what the events of the past year have revealed when it comes to machine learning ethics. #AmazonScholars amazon.science/latest-news/a…
Amazon Scholars Michael Kearns and Aaron Roth discuss the ethics of machine learning

Amazon Scholars Michael Kearns and Aaron Roth discuss the ethics of machine learning

Two of the world’s leading experts on algorithmic bias look back at the events of the past year and reflect on what we’ve learned, what we’re still grappling with, and how far we have to go.

amazon.science
12
2
46
Next talk! "Audience Engagement API: A Privacy Preserving Data Analytics System at Scale", from LinkedIn. #pepr20
1
1
2
Why Privacy Needs Composition ift.tt/3asZvs6
Why Privacy Needs Composition

Why Privacy Needs Composition

We’re back! In our last post we discussed some of the subtle pitfalls of formulating the assumptions underlying average-case relaxations of differential privacy. This time we’re going to look at the...

differentialprivacy.org
5
4
35
Really excited to launch differentialprivacy.org with awesome collaborators @ccanonne_ @JinshuoD @mgtjoseph @thegautamkamath Janardhan Kulkarni Audra McMillan @RowRyRo Adam Sealfon @shortstein and @zstevenwu! Please get in touch if you want to contribute!
1
14
1
48
Our new site, DifferentialPrivacy.org! Hub for all members (theory, practice, anywhere between) of the differential privacy community. Ft learning resources, technical content, and important updates. Community participation invited. Follow @DiffPriv to not miss out. Please RT!🙏
Differential Privacy (@DiffPriv) profile pictureDifferential Privacy@DiffPriv17 Jul 2020
Welcome to DifferentialPrivacy.org! differentialprivacy.org/welc…
3
36
2
77
Today at #MSBuild, I’m proud to announce Microsoft’s #WhiteNoise differential privacy platform is live and ready for developers, researchers and companies worldwide to test, build and support.
2
11
1
21
Latest data from ONS 30 mins ago ons.gov.uk/peoplepopulationa… Just drawn this extraordinary, and shocking, graph. Why the massive rise in non-covid deaths? How much is judgement of registering doctor? How much is collateral damage of lockdown???
Image from David Spiegelhalter (@d_spiegel)
249
1,665
431
2,365
While protecting your privacy, we are sharing aggregated mobility data from Apple Maps to help public health authorities learn how people travel in their communities and to provide valuable insights to stop the spread of COVID-19. Stay safe and healthy! apple.com/newsroom/2020/04/a…
Apple makes mobility data available to aid COVID-19 efforts

Apple makes mobility data available to aid COVID-19 efforts

Apple today released a mobility data trends tool from Apple Maps to support the global work to mitigate the spread of COVID-19.

apple.com
225
825
101
3,907
1/ Last night, I spent some quality time with the Apple docs on the new contact tracing protocol and APIs they and Google are preparing. I'm quite optimistic about this effort. Here's why.
104
1,243
395
3,251
Specs for the Apple-Google Privacy preserving contact tracing are apple.com/covid19/contacttra… It’s privacy-centric, distributed, interoperable and puts users in control
4
12
ICML 2020 (icml.cc/Conferences/2020) will be a virtual conference. We've been having discussions with ICLR folks about their plans, and will be learning from their experience (they are up first). We hope to enable as much of the normal ICML experience as possible, virtually.
7
256
23
726
Scroll up and down for an interesting discussion about concerns re: differential privacy at Census with @davidshor. Bonus points if you can overcome Twitter's terrible thread management and find interesting subthreads by @shortstein and @machanavajjhala!
This quoted post is unavailable.
Try viewing anyway · View on X
5
27
Load more