@elliewhite42.bsky.social
Reposted
New paper's out! First result from excellent DPhil student Brian Rogers - machine learning for scientific discovery. (I'll post a thread explaining in a bit). arxiv.org/abs/2511.09433
What We Don't C: Representations for scientific discovery beyond VAEs
Accessing information in learned representations is critical for scientific discovery in high-dimensional domains. We introduce a novel method based on latent flow matching with classifier-free guidan...
arxiv.org
November 13, 2025 at 4:24 PM
New paper's out! First result from excellent DPhil student Brian Rogers - machine learning for scientific discovery. (I'll post a thread explaining in a bit). arxiv.org/abs/2511.09433