Lightnews — Scholar-powered news

Reposted by Berk Ustun

Loris D'Antoni @lorisdanto.bsky.social · 16d

Who teaches an undergraduate principles of programming languages class? Looking for some inspiration to teach one at UCSD

2 1 5

Reposted by Berk Ustun

Ben Recht @beenwrekt.bsky.social · Sep 8

In a new paper, I try to resolve the counterintuitive evidence of Meehl’s “clinical vs statistical prediction” problems: Statistics only wins because the game is rigged.

The Actuary's Final Word on Algorithmic Decision Making

Paul Meehl's foundational work "Clinical versus Statistical Prediction," provided early theoretical justification and empirical evidence of the superiority of statistical methods over clinical judgmen...

arxiv.org

3 9 28

Berk Ustun @berkustun.bsky.social · Aug 1

Time for XAI for Code? 🙃

1

Reposted by Berk Ustun

Connor Lawless @lawlessopt.bsky.social · Jul 14

Machine learning models can assign fixed predictions that preclude individuals from changing their outcome. Think credit applicants that can never get a loan approved, or young patients that can never get an organ transplant - no matter how sick they are!

1 1 1

Reposted by Berk Ustun

Connor Lawless @lawlessopt.bsky.social · Jul 14

Excited to be chatting about our new paper "Understanding Fixed Predictions via Confined Regions" (joint work with @berkustun.bsky.social, Lily Weng, and Madeleine Udell) at #ICML2025!

🕐 Wed 16 Jul 4:30 p.m. PDT — 7 p.m. PDT
📍East Exhibition Hall A-B #E-1104
🔗 arxiv.org/abs/2502.16380

Understanding Fixed Predictions via Confined Regions

Machine learning models can assign fixed predictions that preclude individuals from changing their outcome. Existing approaches to audit fixed predictions do so on a pointwise basis, which requires ac...

arxiv.org

1 3 5

Reposted by Berk Ustun

Jessica Hullman @jessicahullman.bsky.social · Jul 2

Paper: www.arxiv.org/abs/2506.22740

Blog post: statmodeling.stat.columbia.edu/2025/07/02/w...

Explanations are a means to an end

Modern methods for explainable machine learning are designed to describe how models map inputs to outputs--without deep consideration of how these explanations will be used in practice. This paper arg...

www.arxiv.org

2 3 23

Reposted by Berk Ustun

Jessica Hullman @jessicahullman.bsky.social · Jul 2

ExplainableAI has long frustrated me by lacking a clear theory of what an explanation should do. Improve use of a model for what? How? Given a task what's max effect explanation could have? It's complicated bc most methods are functions of features & prediction but not true state being predicted 1/

2 8 44

Reposted by Berk Ustun

Maria De-Arteaga @mariadearteaga.bsky.social · Jun 25

Having a lot of FOMO not being able to be in person at #FAccT2025 but enjoying the virtual transmission 💻. Tomorrow Jakob will be presenting our paper "Perils of Label Indeterminacy: A Case Study on Prediction of Neurological Recovery After Cardiac Arrest".

screenshot of title and authors (Jakob Schoeffer, Maria De-Arteaga, Jonathan Elmer)

1 1 13

Berk Ustun @berkustun.bsky.social · Jun 24

Explanations don't help us detect algorithmic discrimination. Even when users are trained. Even when we control their beliefs. Even under ideal conditions... 👇

Julian Skirzynski @jskirzynski.bsky.social · Jun 24

Right to explanation laws assume explanations help people detect algorithmic discrimination.

But is there any evidence for that?

In our latest work w/ David Danks @berkustun, we show explanations fail to help people, even under optimal conditions.

PDF shorturl.at/yaRua

10

Berk Ustun @berkustun.bsky.social · May 24

*wrapfig entered the document*

1 1 5

Reposted by Berk Ustun

Peter Sokol-Hessner @p1sh.bsky.social · Apr 26

“Science is a smart, low cost investment. The costs of not investing in it are higher than the risk of doing so… talk to people about science.” - @kevinochsner.bsky.social makes his case to the field #sans2025

7 55 170

Berk Ustun @berkustun.bsky.social · Apr 24

I tried to be nice but then they said that saying please and thanks costs millions.

4

Reposted by Berk Ustun

sorelle @friedler.net · Apr 24

Hey AI folks - stop using SHAP! It won't help you debug [1], won't catch discrimination [2], and makes no sense for feature importance [3].

Plus - as we show - it also won't give recourse.

In a paper at #ICLR we introduce feature responsiveness scores... 1/

arxiv.org/pdf/2410.22598

Left: a feature-highlighting explanation generated by SHAP that shows multiple important features, however these include features that can not be changed (e.g., age, number of dependents) and features that even if they were changed would not result in a different outcome (e.g., credit utilization).

Right: a feature-highlighting explanation generated by our responsiveness scores showing only features that can be changed and which have the potential to result in a better outcome for the individual (multiple credit lines and monthly income).

3 8 29

Reposted by Berk Ustun

Hailey Joren @haileyjoren.bsky.social · Apr 24

When RAG systems hallucinate, is the LLM misusing available information or is the retrieved context insufficient? In our #ICLR2025 paper, we introduce "sufficient context" to disentangle these failure modes. Work w Jianyi Zhang, Chun-Sung Ferng, Da-Cheng Juan, Ankur Taly, @cyroid.bsky.social

1 5 11

Reposted by Berk Ustun

Harry Cheon @scheon.com · Apr 24

Denied a loan, an interview, or an insurance claim by machine learning models? You may be entitled to a list of reasons.

In our latest w @anniewernerfelt.bsky.social @berkustun.bsky.social @friedler.net, we show how existing explanation frameworks fail and present an alternative for recourse

1 7 16

Reposted by Berk Ustun

Karl Rohe @karlrohe.bsky.social · Apr 19

Absolute banger.

Ben Recht @beenwrekt.bsky.social · Apr 19

I have a theory. arxiv.org/abs/2501.03457

A Bureaucratic Theory of Statistics

This commentary proposes a framework for understanding the role of statistics in policy-making, regulation, and bureaucratic systems. I introduce the concept of "ex ante policy," describing statistica...

arxiv.org

4 5 32

Reposted by Berk Ustun

Sujay Nagaraj @snagaraj.bsky.social · Apr 19

Many ML models predict labels that don’t reflect what we care about, e.g.:
– Diagnoses from unreliable tests
– Outcomes from noisy electronic health records

In a new paper w/@berkustun, we study how this subjects individuals to a lottery of mistakes.
Paper: bit.ly/3Y673uZ
🧵👇

1 2 12

Reposted by Berk Ustun

Sujay Nagaraj @snagaraj.bsky.social · Apr 13

🚨 Excited to announce a new paper accepted at #ICLR2025 in Singapore!

“Learning Under Temporal Label Noise”

We tackle a new challenge in time series ML: label noise that changes over time 🧵👇

arxiv.org/abs/2402.04398

Learning under Temporal Label Noise

Many time series classification tasks, where labels vary over time, are affected by label noise that also varies over time. Such noise can cause label quality to improve, worsen, or periodically chang...

arxiv.org

1 1 5

Berk Ustun @berkustun.bsky.social · Feb 28

is this a rhetorical question?

5

Berk Ustun @berkustun.bsky.social · Feb 28

1 1

Reposted by Berk Ustun

Mark Riedl @markriedl.bsky.social · Feb 5

The CHI Human-Centered Explainable AI Workshop is back!

Paper submissions: Feb 20

hcxai.jimdosite.com

Home | HCXAI

ACM CHI 2025 Workshop on Human-Centered Explainable AI (HCXAI). May 2025 (Yokohama, Japan & hybrid). Submit your Paper (EasyChair)

hcxai.jimdosite.com

12 27

Reposted by Berk Ustun

logan koepke @jlkoepke.bsky.social · Jan 21

🧵on the CFPB and less discriminatory algorithms.

last week, in its supervisory highlights, the Bureau offered a range of impressive new details on how financial institutions should be searching for less discriminatory algorithms.

1 5 16

Berk Ustun @berkustun.bsky.social · Jan 14

Also

Reposted by Berk Ustun

Jacobs Center for Health Innovation @UC San Diego Health @jchi-ucsd.bsky.social · Jan 8

Engaging discussions on the future of #AI in #healthcare at this week's ICHPS, hosted by @amstatnews.bsky.social.

JCHI's @kdpsingh.bsky.social shared insights on the safety & equity of #MachineLearning algorithms and examined bias in large language models.

1 4 9

Berk Ustun @berkustun.bsky.social · Jan 7

I imagine it would look like the modern version of this.

For a Year, They Lived Tied Together with 8 Feet of Social Distance

Artists Linda Montano and Tehching Hsieh The promise was made on American Independence Day, 1983. "We, Linda Montano and Tehching Hsieh, plan to do a one year performance. We will stay together for...

www.messynessychic.com

1