Ali Behrouz

alibehrouz.bsky.social

15 January 2025 07:49

🌟 Best of Both Worlds! ❓Have you ever wondered why hybrid models (RNN + Transformers) are powerful? We answer this through the lens of circuit complexity and graphs! Excited to share our work on understanding Graph Sequence Models (GSM), which allows the use of any sequence model for graphs.

Jane Qiu, PhD

janeqiu.bsky.social

13 January 2025 19:22

How does the brain avoid ‘catastrophic forgetting’, the distortion and overwriting of old memories when new ones are created? New clues emerged through a cool method called optogenetics. The insights could inspire solutions to help AI overcome similar challenges. @nature.com 🔗: shorturl.at/mU9GV

Raffaello Camoriano

raf-camo.eurosky.social

13 January 2025 19:22

🖥️🦾 We are pleased to announce that our paper "Long-Term Upper-Limb Prosthesis Myocontrol via High-Density sEMG and Incremental Learning" has been published in the November 2024 issue of the IEEE Robotics and Automation Letters.

Glen Berseth

glenberseth.bsky.social

11 January 2025 21:07

After reading over more NeruIPS papers from this year, one developing realization is that deep #reinforcementlearning may be suffering from a far greater continual learning problem than expected. How much experience is ignored because the model does not have the plasticity to compress new data?

Maria Sofia Bucarelli

mariasofiab.bsky.social

23 December 2024 16:20

🚀Calling ML on Graphs Enthusiasts! I'm co-organizign a Competition on Learning with Noisy Graph Labels at IJCNN 2025. Submissions are now open sites.google.com/view/learnin... Submission Deadline: 10/02/2025 We can’t wait to see your creative approaches and groundbreaking ideas.✨