Understandable Machine Intelligence Lab: We bring #explainable #AI to the next level. Part of @LeibnizATB, Ex @TUBerlin, funded by @BMBF_Bund #XAI
Berlin, Germany
Joined November 2020
- Tweets186
- Following123
- Followers638
- Likes239
Understandable Machine Intelligence Lab retweeted
Happy to share that our PRISM paper has been accepted at #NeurIPS2025 🎉
In this work, we introduce a multi-concept feature description framework that can identify and score polysemantic features.
📄 Paper: arxiv.org/abs/2506.15538
#NeurIPS #MechInterp #XAI
Understandable Machine Intelligence Lab retweeted
If you're at #AAAI2025 don't miss our poster today (alignment track)!
Paper 📘: arxiv.org/pdf/2502.15403
Code 👩💻: github.com/annahedstroem/eva…
Team work with @eirasf and @Marina_MCV
At 12:30 I'll be happy to take questions about our poster presentation at #AAAI2025. Is your explanation for a model's prediction better than the alternatives? "Evaluate with the Inverse: Efficient Approximation of Latent Explanation Quality Distribution" introduces QGE... 1/4
Understandable Machine Intelligence Lab retweeted
At 12:30 I'll be happy to take questions about our poster presentation at #AAAI2025. Is your explanation for a model's prediction better than the alternatives? "Evaluate with the Inverse: Efficient Approximation of Latent Explanation Quality Distribution" introduces QGE... 1/4
Understandable Machine Intelligence Lab retweeted
If you’re still @NeurIPSConf and are curious about how evaluation outcomes of interpretability methods can be adversarially attacked?
👉 Sun, 15 Dec, 4 PM (East Ballroom A and B)
The Price of Freedom: An Adversarial Attack on Interpretability Evaluation
Understandable Machine Intelligence Lab retweeted
Interested in eval, vision and mechanistic interpretability?
Come chat at our @NeurIPSConf poster! #3107
👉 Thu, 12 Dec, 11 AM (East Exhibit Hall A-C)
CoSy: Evaluating Textual Explanations of Neurons
openreview.net/pdf?id=R0bnWr…
Understandable Machine Intelligence Lab retweeted
I’ll be presenting our work at @NeurIPSConf in Vancouver! 🎉
Join me this Thursday, December 12th, in East Exhibit Hall A-C, Poster #3107, from 11 a.m. PST to 2 p.m. PST. I'll be discussing our paper “CoSy: Evaluating Textual Explanations of Neurons.”
Understandable Machine Intelligence Lab retweeted
I am not attending #NeurIPS2024, but I encourage everyone interested in #XAI and #MechInterp to check out our paper on evaluating textual descriptions of neurons!
Join @lkopf_ml, @anna_hedstroem, and @Marina_MCV on Thu 09.12, 1 p.m. to 4 p.m. CST at East Exhibit Hall A-C #3107!
Understandable Machine Intelligence Lab retweeted
✨Introducing quanda: an open-source library for benchmarking training data attribution (TDA) methods in PyTorch!
Quanda offers a user-friendly interface for ⚖️ evaluating attributions and 📊 benchmarking TDA methods across diverse metrics.
Understandable Machine Intelligence Lab retweeted
We (@di_lya, Galip Ümit Yolcu, Niklas Schmolenski, @wiegand_t, @WojciechSamek, @SLapuschkin + me) just launched QUANDA — a training data attribution TDA software
Built for researchers curious to apply/ develop/ evaluate TDA methods
GitHub repo: github.com/dilyabareeva/quan…
Understandable Machine Intelligence Lab retweeted
Thank you Stefan Lindow @SlineL for the amazing opportunity to give a talk on Explainable AI at the Potsdam Graduate School AI for Academy program! Grateful for the insightful audience and great questions! ☺️
Understandable Machine Intelligence Lab retweeted
🎉 Excited to announce that our paper has been accepted to #NeurIPS2024! This is my first first-author publication 🥳 I'm incredibly grateful to my amazing supervisor @kirill_bykov and co-authors @BommerPhiline @anna_hedstroem @SLapuschkin @Marina_MCV!
📄arxiv.org/abs/2405.20331
Understandable Machine Intelligence Lab retweeted
Thats us, the Data Science team at @LeibnizATB! And we are searching 2 new PhD students - If you're passionate about research & coding, become part of our team and make an impact🌍💻
1️⃣ PhD for AI & Bias Detection
2️⃣ PhD for AI & Soil Quality
Apply here: atb-potsdam.de/en/career/vac…
Understandable Machine Intelligence Lab retweeted
Join us today at the #ICML2024 Workshop on the Next Generation of AI Safety! Find @kirill_bykov and me in Hall A1 at Poster Session #2, from 3:30 PM to 4:30 PM. Looking forward to seeing you there!
Understandable Machine Intelligence Lab retweeted
It was an amazing session — great presentations, sparkling questions and fruitful atmosphere.
Pleased to see so many people interested in Global and Concept-Based Explainability!
Join UMI Lab at #XAI2024! 🎉 On July 17th, catch @kirill_bykov at 11:30 chairing the Global Concept-based Explainability track. Then, at 16:00, join @anna_hedstroem for her presentation on "A Fresh Look at Sanity Checks for Saliency Maps". See you in Malta! 🇲🇹 #ExplainableAI
Understandable Machine Intelligence Lab retweeted
Our latest paper is out!
On Tuesday, 17 July, we will present our latest work "A Fresh Look at Sanity Checks for Saliency Maps" at the 2nd XAI World Conference in Valetta, Malta.
I hope to see you there!
Read the paper: link.springer.com/chapter/10…
Code: github.com/annahedstroem/san…
New paper alert!
With EMPRT and SMPRT we fix some glaring issues with the original Model Parameter Randomization Test (MPRT).
Visit our talk at the 2nd XAI World Conference in Valetta, Malta, on July 17th!
more info & paper: tinyurl.com/lnkdn-smprt-empr…
Understandable Machine Intelligence Lab retweeted
Tomorrow is an exciting day
Alongside incredibly talented researchers, I’ll be giving a talk at the UN’s AI for Good summit in Geneva
aiforgood.itu.int/event/expl…
Tune in if you can!
Understandable Machine Intelligence Lab retweeted
Join us next week as @BommerPhiline and I talk about evaluation-centric interpretability in climate science
🗓️ Feb 21, 7am PT / 4pm CET Sign-up (for free)
climatechange.ai/webinars
@ClimateChangeAI
Understandable Machine Intelligence Lab retweeted
Tuesday, January 9th, I'll be presenting our latest #NeurIPS2023 paper "Labeling Neural Representations with Inverse Recognition" at BLISS. Come join!
🗓️ January 9th, 19:15
📍 Location: WiWi Café, EB302, Technische Universität Berlin
📑 Website: bliss.berlin/blog/09-01-24-k…
Understandable Machine Intelligence Lab retweeted
If you are at #NeurIPS2023 check out our poster — happening now at Great Hall & Hall B1+B2 (level 1) #1519!
I am not there — but my co-authors, and @anna_hedstroem from @UMI_Lab_AI will be there 😊
Excited to share our latest #NeurIPS2023 paper "Labeling Neural Representations with Inverse Recognition." (arxiv.org/abs/2311.13594)
We present INVERT — new global explanation method that explains neurons with human-understandable concepts they detect!
A thread 🧵(1/8)
