thermans.bsky.social
Professor of Computing, I do research on robotics and AI. I also ski.
https://robot-learning.cs.utah.edu/thermans
30 posts
168 followers
241 following
Regular Contributor
Active Commenter
comment in response to
post
Big if true!
comment in response to
post
We also have bunnies in our yard!
comment in response to
post
This was supposed to say “provably” not probably. Baby slapping the screen while typing is the likely culprit here.
comment in response to
post
Planning generalizes probably. Yes you need models, it can be expensive, and reactivity can be hard, but we should use it where we can. It’s always funny to me when folks argue for learning over planning for generalization. Yes we need to learn, but planning isn’t whats stopping generalization.
comment in response to
post
I sat next to some candy makers on my flight back from Amsterdam a couple of weeks ago. If I had known their clout with the legislature, I would have asked different questions.
comment in response to
post
And most of our ROS code is barely used by others. The software stack is just too big to easily use someone else’s code.
comment in response to
post
LOL then I would recommend @citationspod.bsky.social
comment in response to
post
Not a podcast but Trevor Noah reading his autobiography "Born a Crime" is a great road trip listen.
comment in response to
post
Travis Deyle wrote a nice article about this in 2012, but his blog seems to be offline, this references it robohub.org/hizook-on-vi...
I also think we shouldn't think of teleop as an upper bound, it definitely shows ability, but teleop is hard and my robots can do things I couldn't teleop.
comment in response to
post
A travesty indeed.
comment in response to
post
Congrats and welcome to the club! I just put in a purchase request which should spend the last dollar of my fellowship money.
comment in response to
post
This is the first suggested use of robot dancing that I can really see being worthwhile.
comment in response to
post
Very much my experience during the Academic Senate Executive Committee today. It would be a lot cooler if my supporters danced behind me though.
comment in response to
post
Thanks for letting me know my paper was accepted!
comment in response to
post
We lost our one short neck giraffe at the grocery store a few weeks ago. I thought it was a cow at first until Kelly corrected me through a large amount of laughter.
comment in response to
post
That's probably not the best reference, just the first one I could find, while holding a sick baby. I'm sure @markguzdial.bsky.social would know the most appropriate.
comment in response to
post
This reminds me a bit of the work on teaching computational thinking to college students and how it doesn't magically transfer to other domains, but transfer can happen if explicitly taught as a goal. E.g. search.app/3aBgfh9h9s5Y...
comment in response to
post
Colleagues who were supposed to be on panels say NSF needs to come up with a procedure to comply with the anti-DEI orders. So how do you hold a panel and not discuss broadening participation and similar things that are currently required by CISE.
comment in response to
post
Behind the scenes making this post: my sick one year old was crawling all over me while I tried to write it on my phone. Also this:
comment in response to
post
@cpaxton.bsky.social discussed this in a blog post not too long ago if you want a tl;dr: itcanthink.substack.com/p/paper-note...
comment in response to
post
We're working on something, I hadn't thought all the way to pup, but let's talk.
comment in response to
post
Most Wanted letterboxd.com/film/most-wa...