Super excited about this work - Using conformal prediction and GANs to meaningfully communicate uncertainty! Anastasios Nikolas Angelopoulos Stephen Bates Yaniv Romano Phillip Isola Presenting in-person at #NeurIPS2022, Thursday @ 4pm.

Interested in conformal prediction after Emmanuel Candès’ NeurIPS keynote? Check out the Gentle Introduction to Conformal Prediction! arxiv.org/pdf/2107.07511… Practical explanations ✅ Code ✅ Real-data examples ✅ Learn to put rigorous UQ on your AI/ML/DL algorithm!

Interested in generative models + uncertainty quantification? Check out this article by MIT CSAIL about our recent work! bit.ly/3HdDbDE "We’d like to know just how uncertain our uncertainty is" - One step closer to trustworthy ML deployments! Trustworthy ML Initiative (TrustML)

We need more nuanced discussions around the risk of open sourcing models. Open source brings valuable access, but it is absurd to ignore the fact that it lowers the barriers to entry for both useful use cases and potential misuse.

CFP is out for our workshop!! Exciting speaker list coming up soon. Please submit and join us at Hawaii!! Note: We'll try our best to expedite reviews for folks who need a VISA!

Join us at ICML'23 to discuss deployment challenges for Generative AI with this amazing set of invited speakers! CFP: tinyurl.com/deployinggener… Deadline: May 19th, 2023 Anywhere on Earth (AoE) Notifications: June 23rd, 2023 Workshop date: July 28, 2023

📢 ICML'23 workshop Deadline extended! 📢 Name: Challenges in Deployable Generative AI Deadline: M̶a̶y̶ ̶1̶9̶t̶h̶ May 26th, 2023 (AoE) CFP: tinyurl.com/deployinggener…

As generative AI models get large, we need to deploy them with increased efficiency and transparency, WITHOUT updating these giant models. GRACE is a first step in this direction; the result of an amazing collaboration! Do check out our paper and code! Trustworthy ML Initiative (TrustML)

I will be at ICML Conference next week. Happy to chat about building responsible generative AI. 1. I'm organizing the workshop on Generative AI deployment, we have exciting content. 2. Starting this Fall, I'll be joining the AI ethics team Sony AI. Excited about this journey!

Personalized models often use group attributes like sex/age/race. In our latest w Berk Ustun Marzyeh, we show how personalization can lead to worsenalization by reducing performance for some groups PDF bit.ly/43yump #ICML23 Oral bit.ly/3Kck7rQ

📢Please RT!📢 I'm recruiting a postdoc to work on Editing and Debiasing LLMs at UVA in collaboration with Steven L. Johnson (he/him) (UVA) and Maarten Sap (he/him) (CMU)! More info on my website: tomhartvigsen.com/join Come chat with me about this at #NeurIPS2023!

This was a great work to be part of. Please visit our poster during Session 1! #ICLR2024

I've recently joined Lila Sciences as a Senior Researcher. I'm truly excited about the possibility of creating agentic interfaces aiding autonomous scientific experimentation. Excited to work with Andrew Beam and the rest of the team!

If you're at Neurips next week, come hang out with us and learn how our team Lila Sciences is transforming autonomous scientific experimentation!