Profile avatar
mniepert.bsky.social
Professor @ University of Stuttgart, Scientific Advisor @ NEC Labs, GraphML, geometric deep learning, ML for Science and Simulations. Formerly @IUBloomington and @uwcse
8 posts 1,172 followers 355 following
Regular Contributor
comment in response to post
[9/n] Beyond Image Generation LD3 can be applied to diffusion models in other domains, such as molecular docking.
comment in response to post
We analyzed the behavior of Gumbel softmax in complex stochastic computation graphs. It’s a combination of vanishing gradients and a tendency to fall into poor local minima, underutilizing available categories. We also have some ideas for improvements. proceedings.neurips.cc/paper/2021/h...
comment in response to post
🙋‍♂️
comment in response to post
🙋‍♂️
comment in response to post
But yeah: reviewing for ICLR now and that’s also good to have more senior people in the mix there.
comment in response to post
Don’t know about SAC but as AC I typically have several borderline papers where I have to write an additional review and poke reviewers/authors to address certain points I might disagree with. For last NeurIPS I went through all borderline papers with SAC on Zoom.
comment in response to post
I also played with the same bb multiple pole implementation a few months ago, hoping to build an end-to-end learnable multipole MPNN. So learn a kernel function for 3d point clouds to model long-range interactions. Unfortunately I had to give up due to lack of time and the complex code. :)