Michael "Shapes Dude" Betancourt
@betanalpha.bsky.social
Zealous modeler. Annoying statistician. Reluctant geometer. Support my writing at http://patreon.com/betanalpha. He/him.
Most people read papers like this, which is why captions are critical to engaging with your audience. Captions need to be self-contained, explaining why the figure/plot is meaningful. (Another insight from www.principiae.be/X0100.php).
Trees, maps, and theorems
Trees, maps, and theorems
principiae.be
Angela Collier@acollierastro.bsky.social · 3d agoEvery time I open a paper I just scroll til I hit plots and then I feel like a bad scientist because I will only read the text if the plots interest me. Can't stop won't stop.
Most people read papers like this, which is why captions are critical to engaging with your audience. Captions need to be self-contained, explaining why the figure/plot is meaningful. (Another insight from www.principiae.be/X0100.php).
Trees, maps, and theorems
Trees, maps, and theorems
principiae.be
Angela Collier@acollierastro.bsky.social · 3d agoEvery time I open a paper I just scroll til I hit plots and then I feel like a bad scientist because I will only read the text if the plots interest me. Can't stop won't stop.
After (mostly) recovering from the scars of drafting the conditional probability theory chapter a few years ago, I'm back in the probability theory game with my next chapter covering probability theory applied on product spaces.
New Probability Theory On Product Spaces Chapter | Michael Betancourt
New Probability Theory On Product Spaces Chapter by Michael Betancourt on Patreon. Join Michael Betancourt's community for exclusive content and updates.
patreon.com
“The only one who could ever reach me Was the son of a preacher man.” I’m choosing to believe that “Son of a Preacher Man” was about Thomas Bayes.
“The only one who could ever reach me Was the son of a preacher man.” I’m choosing to believe that “Son of a Preacher Man” was about Thomas Bayes.
the strait of hormel, also known as the spam filter
After (mostly) recovering from the scars of drafting the conditional probability theory chapter a few years ago, I'm back in the probability theory game with my next chapter covering probability theory applied on product spaces.
New Probability Theory On Product Spaces Chapter | Michael Betancourt
New Probability Theory On Product Spaces Chapter by Michael Betancourt on Patreon. Join Michael Betancourt's community for exclusive content and updates.
patreon.com
A nice exercise in appreciating good approximations versus bad approximations. <glares at statistical computation>
Ayliean@ayliean.bsky.social · 6d agoYes there’s a solar eclipse and a lunar eclipse coming up this month… but are you ready for 🥧 “Not Quite As Good Pi Approximation Day” 🥧 on 25/8?!
A nice exercise in appreciating good approximations versus bad approximations. <glares at statistical computation>
Ayliean@ayliean.bsky.social · 6d agoYes there’s a solar eclipse and a lunar eclipse coming up this month… but are you ready for 🥧 “Not Quite As Good Pi Approximation Day” 🥧 on 25/8?!
My conspiracy theory is that the accumulation of talent to the Dodgers is collusion by the league to try to sell fans on a salary cap...
Using lists of three is not an "AI tell." It's a common trait in human writing because our brains biologically want to make sure there's always one thing to marry, one thing to fuck, and one thing to murder.
Harrison Ford, the actor behind loveable scoundrels Han Solo and Indiana Jones, started his career as an on-set carpenter. That makes him a roguelike deckbuilder.
Conjugated myself into the wrong coset and missed my friends. Now I'm getting roasted in the quotient group chat
Friendly reminder that everyone is welcome to join the probabilistic modeling Discord server to discuss all things probabilistic modeling, www.patreon.com/betanalpha/p....
Generative Modeling Discord Server | Michael Betancourt
[Despite my best efforts the invite link can sometimes expire. If you find that the link is no longer working then try clearing your cookie
patreon.com
Friendly reminder that everyone is welcome to join the probabilistic modeling Discord server to discuss all things probabilistic modeling, www.patreon.com/betanalpha/p....
Generative Modeling Discord Server | Michael Betancourt
[Despite my best efforts the invite link can sometimes expire. If you find that the link is no longer working then try clearing your cookie
patreon.com
Now that it has been released I am allowed to share my audition for The Odyssey. My manager says they thought I was too broad and whatnot.
The IgNobel goes to everyone who used a black box frequentist estimator, assuming that the asymptotic calibration applied universally.
Randall Munroe@xkcd.com · 2w agoCalibration Nobel xkcd.com/3275/
![Comic. [Person 1 with short hair on stage behind podium reading off paper while addressing audience. People are shown sitting in chairs while Person 2 with ponytail and Person 3 waving are walking toward stage.] PERSON 1: This year’s Nobel Prize in “calibrating your equipment carefully and not embarrassing yourself by publishing spurious results” goes to Dr. Adams and Dr. Smith, who did *not* announce the discovery of microbes on Titan. PERSON 3: Thank you, thank you.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:cz73r7iyiqn26upot4jtjdhk/bafkreihztdrvxjeyzmz6dmz3xb4vi6wiwwtge6fwdoxjcum5aptndv3kfm)
The IgNobel goes to everyone who used a black box frequentist estimator, assuming that the asymptotic calibration applied universally.
Randall Munroe@xkcd.com · 2w agoCalibration Nobel xkcd.com/3275/
People are drawn to black boxes because they externalize costs. People will choose black boxes, even when more open alternatives exist, because without those costs they are by far the easiest option. To make more open alternatives appealing, there has to be enough friction to use a black box.
People are drawn to black boxes because they externalize costs. People will choose black boxes, even when more open alternatives exist, because without those costs they are by far the easiest option. To make more open alternatives appealing, there has to be enough friction to use a black box.
"Also consider whether non-proprietary, energy-efficient, or small-scale systems suffice for your task. If not, consider how preservation of the values articulated in this Declaration may be worth a delay in obtaining results."
In my recent courses, people have been less and less likely to ask questions, and they usually admit it’s because they don’t want to bother me. The course instructor. In a course designed to maximize opportunities to ask questions. It feels sincere, but it’s definitely a bit of a phase transition.
The Emotion Engine@emotion-engine.bsky.social · 3w agoPeople ask AI because it is, to a fault, agreeable and enthusiastic to “help” you and that comforts the universal fear of being seen as annoying BUT what we often forget is any expert who could actually help you is going to be fucking psyched to get into the weeds with you about their field.
Can't believe all this shit still happened, even when that lady in Maine was expressing mild concern and whatnot. Her mild concern shoulda stopped all this shit in its tracks.
Anyone know of any papers that directly critique Markov chain Monte Carlo methods for being unreliable (and not just slow/slower than a competing method)? Papers in stats or applied fields welcome, and the critiques don't have to be well-formed or even correct.
Adrien Corenflos@adriencorenflos.bsky.social · 3w agoOh I'm actually looking for/collating examples of people (papers) actively shunning MCMC for being unreliable or non-reproducible. Would you happen to know of specific references where this was argued?