🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

We’re thrilled to announce the first competition on unlearning, as part of the #neurips competition track! Joint work w Fabian Pedregosa, Vincent Dumoulin, Gintare Karolina Dziugaite, Ioannis Mitliagkas (Γιάννης Μητλιάγκας), @KurmanjiMeghdad, Peter Triantafillou, Isabelle Guyon and others. Read our blog post!

📢The NeurIPS 2023 Unlearning Competition is now open for submissions Kaggle !📢 Goal is to develop algorithms that can unlearn a subset of the training data. Competition is open to all, and there are prizes for the top performers!. kaggle.com/competitions/n…

Soooo excited that our unlearning competition is now live and accepting submissions 🎉 Can’t wait to see what unlearning algorithms you come up with!!

cacm.acm.org/blogs/blog-cac…

The NeurIPS unlearning-challenge Competition is open for submissions Kaggle→ goo.gle/45ONCc1! With the goal of developing algorithms that can unlearn a subset of training data, the competition is open to all, with prizes for the top performers. More from our earlier blog ↓

Funny how some folks who think theory has some magical properties readily dismiss bona fide engineering and empirical science as alchemy. Blind trust in theoretical results that turned out to be irrelevant is a major reason why neural nets were dismissed between 1995 and 2010.

Two weeks into the unlearning challenge and there are already 400 submissions! Now’s the perfect time to join the challenge and help develop new machine learning algorithms that can safely and effectively unlearn outdated or harmful information. kaggle.com/competitions/n…

Our paper on #machineunlearning will be presented at #NeurIPS2023 . In this paper, we introduce SCRUB, a method to efficiently unlearn from deep models at scale. Check out the latest preprint!

Can LLMs really self-critique (and iteratively improve) their solutions, as claimed in the literature?🤔 Two new papers from our group investigate (and call into question) these claims in reasoning (arxiv.org/abs/2310.12397) and planning (arxiv.org/abs/2310.08118) tasks.🧵 1/

I am more interested in a system with the cognitive abilities of a 4-year-old than in a big curve that has memorized the entire Internet.

1/ What makes machine unlearning hard? And what can we do about it? Our #NeurIPS2024 paper explores key factors affecting unlearning difficulty & introduces RUM🍹, a new framework to improve unlearning pipelines - w/Meghdad Kurmanji, George Barbulescu, Eleni Triantafillou,

The Cambridge ML Systems Lab (Cambridge ML Systems Lab), working together with Flower, has been selected as one of a handful of teams from around the world to invent open-source decentralized ML solutions in a program funded by the German government. This farsighted initiative developed by

This will be fun! Great speakers for the Cambridge ML Systems Lab talk series, thanks to Bill Marino and Meghdad Kurmanji.