Starting day-2 of CoLLAs with an excellent keynote on continual reinforcement learning by Doina Precup!

Unsupervised + federated + continual 🤩 If you are attending CoLLAs 2025 #CoLLAs2024 please come check out our work “Masked Autoencoders are Efficient Continual Federated Learners” in today’s (Tuesday) poster session at 15:30-18:00! arxiv.org/abs/2306.03542 W/ Kristian Kersting

#CoLLas2024 folks 🐨 Temperature rising for the coming sessions 🔥🐨

I’m definitely biased. But CoLLAs 2025 is really the best ML conference to attend if you care about developing true intelligent systems. Continual learning is the grand problem we need to solve before worrying about AGI killing us all. Plus you will not hear about AGI nonsense

#CoLLas2024 The fifth keynote by Jay McClelland Is Just starting! 🐨

#CoLLas2024 Sixth Keynote speach by Eleni Triantafillou 🐨

Be the reviewer you want to see in the world

Apparently it’s now “feminism” to relentlessly bully and put in danger women who don’t look “feminine” enough

I am very excited about this work, which we will present in October! 🇲🇦 ♾ #ContinualLearning methods for medical image segmentation that reply scans from previous training stages often perform the best, yet storing patient data is usually not feasible 👉 Pseudo-rehearsal instead

Can’t believe #CoLLAs2024 is already over. It was AMAZING & I’m very grateful to all organizers, speakers, authors & participants. CoLLAs reminded me once more how incredible science is when the community chooses to push together & discuss constructively. Let’s keep this up ❤️

After Pisa, Italy, for CoLLAs 2024, which just ended a week back, CoLLAs 2025 is moving to the USA for its fourth edition! We are happy to announce that CoLLAs 2025 will be happening at UPenn, USA, with a great team of program chairs! More details to be announced soon!

I’m lost for words when i see the idea of “automating science” being celebrated and presented as the epitome of advancement

In preparation for Data-Centric Dynamic Learning at MICCAI 2024, we're hosting a seminar with the brilliant Jaehong Yoon! 🎓 Join us as he delves into federated continual learning agents! 👉dub.sh/dynamicmedai 🗓️Wed, August 28th 🕔 5 pm CEST 🕗 8 am PT 🕚 11 am EST

🎉 Proceedings of the 1st ContinualAI Unconference are out: proceedings.mlr.press/v249/ Papers followed a two-stage pre-registration process. Very refreshing to read about the positive & negative outcomes! ⭐️24 hours of amazing talks are also recorded here: youtube.com/playlist?list=…

Call for papers for AISTATS 2025 is out now aistats.org/aistats2025/ca… You can also (self-)nominate for reviewer/AC position forms.gle/saediFoTznTQ7h… We are (trying to) switch to OpenReview so submission details will be available a bit later. Stephan Mandt yingzhen Shipra Agrawal 1/

Hey all! Applications are open for the @Outreachy internship in open source for those underrepresented in tech. 💵 $7,000 stipend 🌍 Remote work 🕰 3-month duration (Dec-Mar) Apply by Aug 21st! 👇🏾

You want to efficiently search for architectures & hyperparameters jointly in federated learning w/ privacy guarantees? Then check out our AutoML_conf workshop paper: openreview.net/forum?id=dzHIL… Joint work with Pooja Prasad, Devendra Singh Dhami | देवेंद्र सिंह धामी, Martin Mundt & Kristian Kersting