![Magnus Petersen @Neurips Profile](https://pbs.twimg.com/profile_images/1799578426405208064/bIV_AE6P_x96.jpg)
Magnus Petersen @Neurips
@Omorfiamorphism
Followers
765
Following
977
Statuses
237
Deep Learning methods dev for creativity 👾🖼️| Ph.D. student researching deep learning for molecular dynamic simulations @fias_science @CovinoLab
Frankfurt am Main, Germany
Joined September 2013
RT @martinmbauer: If anything we aren't obsessed enough. Mathematical ignorance should be treated like illiteracy. The fact that a large fr…
0
611
0
RT @diego_gcerdas: ✨Our work on BrainACTIV has been accepted at ICLR 2025!✨ Brain Activation Control Through Image Variation (BrainACTIV)…
0
14
0
RT @lugaricano: Europe needs to wake up NOW: The US races ahead with regulatory flexibility and a huge ($500bn!) investment in AI infrast…
0
122
0
RT @JJitsev: Being centrally involved in two largest EU projects on frontier foundation models (eg, openEuroLLM), I would strongly advice r…
0
8
0
@EkdeepL Amazing work! How is overprompting done in practice for SD1.4? Do you crank up the cross attention for tokens associated with the overprompted concept?
0
0
2
RT @EkdeepL: Paper alert—accepted as a NeurIPS *Spotlight*!🧵👇 We build on our past work relating emergence to task compositionality and an…
0
91
0
RT @martinagvilas: New position paper accepted at #ICML2024! 💻🧠 "An Inner Interpretability Framework for AI Inspired by Lessons from Cogni…
0
57
0
RT @RolandDunbrack: Demis -- I think AlphaFold3 is really exciting. As Reviewer #3, I got great results from the server. I tried hard to ge…
0
372
0
RT @_KarenHao: For years I’ve been interviewing data annotation workers who are the lifeblood of the AI industry. For years I’ve heard the…
0
952
0
RT @zzznah: Kunstformen der Natur + IsoNCA, a tribute to Ernst Haeckel Full 30-pattern video:
0
14
0
Would be interesting to see how the fractal dimension changes as a function of parameter count. Maybe it gets smoother as the count increases or maybe even worse 😅
Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.
0
0
0
RT @jaschasd: Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like…
0
2K
0
RT @martinagvilas: I’ll be presenting this work today at 10.45 in #NeurIPS2023 👇 Drop by or reach out by DM if you are interested in inter…
0
3
0