Rob Patro
@robp.bsky.social
Associate Professor of CS @ University of Maryland. Proud Rust advocate! I ♥ science & compiled, statically-typed programming languages! Views are my own. Tech stack: https://github.com/rob-p/tech-stack.
They're laying it on thick, huh?
November 3, 2025 at 6:27 PM
They're laying it on thick, huh?
Hodge decomp is easy though. It breaks up 1 forms into disjoint components. In this case, perhaps a vector field into a divergence free component, a curl free component, and a harmonic component (pic from Keenan’s course)?
October 24, 2025 at 11:28 AM
Hodge decomp is easy though. It breaks up 1 forms into disjoint components. In this case, perhaps a vector field into a divergence free component, a curl free component, and a harmonic component (pic from Keenan’s course)?
San Diego is pretty nice 🙂!
October 21, 2025 at 2:15 PM
San Diego is pretty nice 🙂!
I'm low-key excited for the term "DNS resolution" to enter the broader public vernacular. I just wish it had a more positive association...
October 20, 2025 at 9:05 PM
I'm low-key excited for the term "DNS resolution" to enter the broader public vernacular. I just wish it had a more positive association...
I find this conference where both the submissions and reviews are AI generated at least a bit distressing. Also, it would seem state of the art generative models still can’t solve the LaTeX formatting problem 😜 agents4science.stanford.edu
October 20, 2025 at 2:30 PM
I find this conference where both the submissions and reviews are AI generated at least a bit distressing. Also, it would seem state of the art generative models still can’t solve the LaTeX formatting problem 😜 agents4science.stanford.edu
Getting ready to watch the BSO play along with Jurassic Park (the OG)! Should be good!
October 5, 2025 at 6:58 PM
Getting ready to watch the BSO play along with Jurassic Park (the OG)! Should be good!
Yes; that's what they believe. But I disagree that publicizing this is a bad use of time. I'm the polarized online world, we often forget how many people are not terminally online or who are not hardline one one "side" or the other. Pointing out rank and obvious hypocrisy is good, actually.
September 18, 2025 at 1:59 PM
Yes; that's what they believe. But I disagree that publicizing this is a bad use of time. I'm the polarized online world, we often forget how many people are not terminally online or who are not hardline one one "side" or the other. Pointing out rank and obvious hypocrisy is good, actually.
We've had a preprint about 80% ready for like 10 months.... That last 20% is killing me as I don't have a student with dedicated time for this project...
September 17, 2025 at 2:26 PM
We've had a preprint about 80% ready for like 10 months.... That last 20% is killing me as I don't have a student with dedicated time for this project...
Oh noes, the AI is getting smarter and coming for us!
The AI that is coming for us:
The AI that is coming for us:
September 7, 2025 at 11:59 PM
Oh noes, the AI is getting smarter and coming for us!
The AI that is coming for us:
The AI that is coming for us:
I have a bunch of examples!
September 6, 2025 at 2:50 PM
I have a bunch of examples!
Certainly not bioinformatics (De Bruijn graph "example" from Claude):
September 5, 2025 at 9:03 PM
Certainly not bioinformatics (De Bruijn graph "example" from Claude):
I asked claude nicely, to come up with a small De Bruijn graph example for me. This is what it came up with (this is after 3 rounds of trying to get it to do better). The AI apocalypse is upon us!
September 5, 2025 at 8:26 PM
I asked claude nicely, to come up with a small De Bruijn graph example for me. This is what it came up with (this is after 3 rounds of trying to get it to do better). The AI apocalypse is upon us!
Yes! Finally, they are beginning to believe.
September 5, 2025 at 2:18 PM
Yes! Finally, they are beginning to believe.
Indeed, ChatGPT & Claude still refer to your 2015 paper :)
September 3, 2025 at 4:33 PM
Indeed, ChatGPT & Claude still refer to your 2015 paper :)
Ughhh, I almost mistook this for hazelnut and ruined a perfectly good coffee!!
August 27, 2025 at 11:06 PM
Ughhh, I almost mistook this for hazelnut and ruined a perfectly good coffee!!
Sorry, but this should be illegal before September!
August 26, 2025 at 1:30 PM
Sorry, but this should be illegal before September!
Without downloading any new images, where are you mentally?
August 26, 2025 at 2:10 AM
Without downloading any new images, where are you mentally?
Last session of #WABI25 kicks off with "Which Phylogenetic Networks are Level-k Networks with Additional Arcs? Structure and Algorithms" presented by Takatora Suzuki.
August 22, 2025 at 2:32 PM
Last session of #WABI25 kicks off with "Which Phylogenetic Networks are Level-k Networks with Additional Arcs? Structure and Algorithms" presented by Takatora Suzuki.
Mateo Gray tells us about the latest and greatest in RNA energy minimization in "Spark: sparsified hierarchical energy minimization of RNA pseudoknots" #WABI25!
August 22, 2025 at 1:36 PM
Mateo Gray tells us about the latest and greatest in RNA energy minimization in "Spark: sparsified hierarchical energy minimization of RNA pseudoknots" #WABI25!
First talk of the last day of #WABI25 "Identifying Breakpoint Median Genomes: A Branching Algorithm Approach" presented by Arash Jamshidpey!
August 22, 2025 at 1:03 PM
First talk of the last day of #WABI25 "Identifying Breakpoint Median Genomes: A Branching Algorithm Approach" presented by Arash Jamshidpey!
"Graph-Based Extension of Partial Atom Maps: Theoretical Uniqueness and Practical Algorithms" being presented by Marcos E. González Laffitte at #WABI25. A topic quite different from many of the others in the conference, but nonetheless very interesting!
August 21, 2025 at 7:03 PM
"Graph-Based Extension of Partial Atom Maps: Theoretical Uniqueness and Practical Algorithms" being presented by Marcos E. González Laffitte at #WABI25. A topic quite different from many of the others in the conference, but nonetheless very interesting!