TrustML-(Un)Limited

Official Twitter account for "Pitfalls of limited data and computation for Trustworthy ML" workshop to be held at ICLR 2023.

[TrustML-(Un)Limited @TMLunLimited]

Thanks everyone for coming to the workshop. We thoroughly enjoyed it and learned a lot.

Hopefully see you again in the next iteration.

[TrustML-(Un)Limited @TMLunLimited]

🚀🚀We are excited to be finally hosting our #ICLR2023 workshop on the effects of resource scarcity on privacy, fairness, robustness. Stop by room #MH2 for talks by @vitalyFM, @fereshte_khani, @pnetrapalli, Nicolas Carlini, Ruth Urner.

[TrustML-(Un)Limited @TMLunLimited]

Now @vitalyFM is discussing memorisation in deep neural networks (and how xkcd artists are ahead of AI researchers)

[TrustML-(Un)Limited @TMLunLimited]

Now we have @PNetrapalli talking about why neural networks are not robust to distribution shift and how we can make them so

[TrustML-(Un)Limited @TMLunLimited]

We started off with @alexandrutifrea giving the opening remarks and highlighting some of the trustworthinees metrics that will be discussed in the workshop!

[TrustML-(Un)Limited @TMLunLimited]

The workshop is co-organised by @AmartyaSanyal @alexandrutifrea @fraboeni @subGaussian @FannyYangETH @sarahookr @ilssontfous @jamiemorgenste1

[TrustML-(Un)Limited @TMLunLimited]

🚀🚀We are excited to be finally hosting our #ICLR2023 workshop on the effects of resource scarcity on privacy, fairness, robustness. Stop by room #MH2 for talks by @vitalyFM, @fereshte_khani, @pnetrapalli, Nicolas Carlini, Ruth Urner.

[TrustML-(Un)Limited @TMLunLimited]

We will be hosting a panel discussion joined by @fereshte_khani @jainprateek_ @sarahookr and Nicholas Carlini.

To get the discussion started, type your answer to

"What are the biggest obstacles to private, robust, and fair machine learning ?" 👇

https://t.co/OfYTFshjQ1

[Amartya Sanyal @AmartyaSanyal]

Excited to be organising the @TMLunLimited at @iclr_conf in just one week. You can find @alexandrutifrea @jainprateek_ @sarahookr in person in Kigali!

Our excellent invited and contributed talk list is now online. https://t.co/w4xW45Yprf

[Mara Graziani @mormontre]

I'll be @iclr_conf next week to present concept discovery with SVD @TMLunLimited. We propose a novel framework that uses Singular Value Decomposition to automatically identify concepts learned by the model at intermediate representations. [1/4]

[TrustML-(Un)Limited @TMLunLimited]

📣 #ICLR2023 is coming up soon in Kigali and our workshop is almost ready to go! 🌍🔬

Check out all the accepted papers at our openreview link 👇 and join us on May 5th for an exciting discussion! #TrustworthyML

https://t.co/Bst57Ho8dJ

[Poinsot Audrey @PoinsotAudrey]

Excited to attend ICLR 2023 with @alessandroleite to present our recent work on causal data augmentation at the Trust-ML-(un)Limited Workshop @TMLunLimited

https://t.co/qOG6I2ktBB

[Mara Graziani @mormontre]

Thanks for inviting me, @ExplainableML! It was a fantastic event!

Here the preprint with @LauraAOMahony @henmueller1 @phineas_zu and @VAndrearczyk:
https://t.co/0txXvd7hBt

Stay tuned for more info and hope to see some of you @TMLunLimited @iclr_conf in marvellous Rwanda!

[TrustML-(Un)Limited @TMLunLimited]

🚀🚀🚀The decisions for the workshop on Pitfalls of limited data and computation for Trustworthy ML (#ICLR2023) have been released.🚀🚀🚀

Congratulations to the authors of all accepted papers! The review process was an intense effort with >50 reviewers. See you at @iclr_conf!

[Vector Institute @VectorInst]

.@fraboeni is a Postdoctoral Fellow at Vector Institute, supervised by Prof. @NicolasPapernot. Franziska specializes in Trustworthy Machine Learning & Privacy with a focus on indiv users & data owners. Learn more here🧵👇

https://t.co/Sy43NlJCHs

#VectorResearcherFridays

[TrustML-(Un)Limited @TMLunLimited]

⏰Deadline Extension⏰

Due to the many incoming requests, the deadline has been extended to Feb 15th noon CET. Keep all those amazing submissions coming.

If you are looking for a rundown of the CfP, it is "barriers to trustworthy ML and algorithms that overcome them"

[TrustML-(Un)Limited @TMLunLimited]

Happy new year everyone!

We are excited to announce that will be accepting submissions starting from 🚀Jan 15, 2023🚀 on https://t.co/Bst57Ho8dJ

⏰ Deadline February 8th, 2023, 11:59 PM AoE

See the full CfP on our website https://t.co/lX2U65XsNt

[TrustML-(Un)Limited @TMLunLimited]

Happy new year everyone!

We are excited to announce that will be accepting submissions starting from 🚀Jan 15, 2023🚀 on https://t.co/Bst57Ho8dJ

⏰ Deadline February 8th, 2023, 11:59 PM AoE

See the full CfP on our website https://t.co/lX2U65XsNt

[TrustML-(Un)Limited @TMLunLimited]

Our Call for Papers is also online. Check it out to see the two different tracks

1. Regular short paper track (6 pages)

2. Exploratory track (4 pages) to show a novel preliminary observation (theoretical or empirical) and encourage collaboration

https://t.co/lX2U65FRoT

[TrustML-(Un)Limited @TMLunLimited]

We will have many interesting invited talks from
@fereshte_khani @thegautamkamath @PNetrapalli, Ruth Urner, @vitalyFM

And finally, our organising team:
@alexandrutifrea @AmartyaSanyal @subGaussian @FannyYangETH @fraboeni @jamiemorgenste1 @jainprateek @sarahookr @ilssontfous

[TrustML-(Un)Limited @TMLunLimited]

Excited to announce our in-person ICLR 2023 workshop on "Pitfalls of limited data and computation for trustworthy ML".

We want to foster discussion on how practically relevant statistical and computational limitations can harm trustworthiness in ML.

https://t.co/PcZX4adoqh

[TrustML-(Un)Limited @TMLunLimited]

Our Call for Papers is also online. Check it out to see the two different tracks

1. Regular short paper track (6 pages)

2. Exploratory track (4 pages) to show a novel preliminary observation (theoretical or empirical) and encourage collaboration

https://t.co/lX2U65FRoT