marcelhussing.bsky.social
PhD student at the University of Pennsylvania. Currently, intern at MSR. Interested in reliable and replicable reinforcement learning and using it for knowledge discovery: https://marcelhussing.github.io/
All posts are my own.
119 posts
2,751 followers
344 following
Regular Contributor
Active Commenter
comment in response to
post
I agree with the notion but I don't think "things being outdated" is always bad. I'm of the opinion that we should still teach SVMs/Kernels as they teach us a different way to think about ML. PCA is still a core tool to teaching low-dim embeddings to students. We need as many tools as possible.
comment in response to
post
Are there no spotlights this year? Do we know?
comment in response to
post
It's called exploratory preference optimization arxiv.org/abs/2405.21046 by @djfoster.bsky.social and others :)
comment in response to
post
This is huge, I might be able to make it now woohoo
comment in response to
post
Wish I could recommend the Kingkiller Chronicle but we may never see an ending. If you are fine with that, the first two been my favorite books for years and I still go back regularly.
comment in response to
post
One key is to follow people and simply engage. Like I spent years on twitter and ended up with 300 followers. Here, I felt much more appreciated. I want to tweet things because people may read it.
comment in response to
post
With great power comes great responsibility 🧐
comment in response to
post
Done!
comment in response to
post
Let's gooooo
comment in response to
post
I do not, will email you sometime later today!
comment in response to
post
We should grab coffee. Claas told me you're cooking something up together.
comment in response to
post
Am I fine to put my name down even if I'm only going on the market next year?
comment in response to
post
But isn't the context switch just distracting?
comment in response to
post
Oh thanks for sharing I wasn't aware of this!
comment in response to
post
Very curious what your thoughts on application areas are! I'll be there, any specific time you'll be around?
comment in response to
post
Everyone better be posting their best theorems or I am removing them all
comment in response to
post
Done!
comment in response to
post
How am I supposed to make February? 😩🫠
comment in response to
post
Just realized I posted the wrong link. The first one was supposed to be this one: arxiv.org/abs/2304.09853
comment in response to
post
Are you talking about empirical benchmarks? Then, I am not aware of much work in this direction other than:
arxiv.org/abs/2410.08870
arxiv.org/abs/1709.06009
@cvoelcker.bsky.social and I arguing that more research need to be done in this area: arxiv.org/abs/2410.08870 and are on it.
comment in response to
post
But if I say, "I just made a post" that does not make clear *where* I posted. Personally, not a fan.
comment in response to
post
Reskeets for visibility are greatly appreciated!
comment in response to
post
Tell them to come here and hit me up go.bsky.app/21nFz12
comment in response to
post
Casually plugging this arxiv.org/abs/2410.08870 we got lots of nodding and "yea of course" for this but nobody seems to want to change it. cc @cvoelcker.bsky.social