rscho

55 ms·

901.
by rscho 6y ago
In response to today's checked C submission. In this video, the speaker decomposes ATS to show that it is really an ML shell over C code.
902.
ATS: Why Linear Types Are the Future of Systems Programming(youtube.com)
2 points by rscho 6y ago|1 comments
903.
by rscho 6y ago
The docs: https://docs.racket-lang.org/guide/languages.html The book: https://beautifulracket.com/ The video: https://youtu.be/WQGh_NemRy4
904.
by rscho 6y ago
I'm not a Kotlin expert, but I expect that it enables easily interpreted DSLs. Racket enables easy compiled DSLs through its macro system. So, no interpretative overhead. Also you can really, really change both syntax and semantics
905.
by rscho 6y ago
In substance: easily make a DSL targeted specifically at your problem. This proposition is often criticized by software engineers for resulting in code that is potentially inscrutable to anyone but the writer. IMO, this misses the forest fo
906.
by rscho 6y ago
I don't know the current memory consumption, but FWIW Racket recently improved on that front. As for debug mode, you should AOT compile with 'raco make' to run a fair test.
907.
by rscho 6y ago
Use C macros long enough and you'll see one of the reasons why Lisp has merit.
908.
by rscho 6y ago
You can not like the language, but frankly: 'Why modernize a crufty Lisp when there is Clojure?' is a strange comment. Those two languages have more differences than they have common ground. "It's got parentheses, and al
909.
by rscho 6y ago
> we don't know how many doctors are relying on these types of recommendations and acting on them I can answer that: close to zero. Clinicians don't want stuff that makes recommendations, as good as they may be. They want a byc
910.
by rscho 6y ago
... which is apparently pure speculation
911.
by rscho 6y ago
That the main problem is not in lab model performance, but in data availability and quality. Most useful applications could be very dumb models pertaining to trivial things. But we need a solid base of quality data for that, that we absolut
912.
by rscho 6y ago
> I know of at least one Canadian hospital that's incorporated ML into 100% of their ED triage. Cool! Which hospital is that? Is the clinical staff happy with the results? Personally, I've never seen any medical ML application
913.
by rscho 6y ago
IMO, those guys really have understood what the clinical world is waiting for regarding medical AI. What is lacking for efficient medical AI is not shiny new models and toys. It's policy changes encouraging the automated gathering of h
914.
MediKanren: A System for Bio-Medical Reasoning(youtube.com)
2 points by rscho 6y ago|1 comments
915.
by rscho 6y ago
We don't know yet. What's sure is that it has greatly increased the number of cancer surgeries, especially lungs. We don't know if that's a good thing yet.
916.
by rscho 6y ago
Regarding medical AI, those guys are pretty much the only ones who seem to have understood what ML can bring to the table TODAY. And that's not even really ML, just logic programming: https://www.uab.edu/medicine/p
917.
by rscho 6y ago
Until they have to really show that it's working in day-to-day practice. Where it most likely won't, unfortunately.
918.
by rscho 6y ago
Yes, if you carry some weight in the field then the most useful contribution would be to push for better automated gathering of quality clinical data. This is the most limiting factor currently. Of course policy activism is far less sexy th
919.
by rscho 6y ago
> This work has greatly increased accuracy in diagnosis, saving lives. As an MD with a special interest in statistics, color me skeptical. I'd love to be proven wrong though, so please provide references. Edit: yeah, so the way this
920.
by rscho 6y ago
> Is there really a huge problem with people overdosing on ibuprofen and suffering permanent damage or death? From personal experience, it would seem that permanent damage is the norm more than it is the exception. But at large I don&#x
921.
by rscho 6y ago
Of course people believed strange things! As I said, the field is far from being hard science even today (am MD myself). What do you think "expert opinion" actually means? Well, just that... and the literature is full of such opin
922.
by rscho 6y ago
TBF, they should stop selling most things OTC. In the current US system that would mean massive additional costs to the consumer, but that is really a policy problem and not a medical one.
923.
by rscho 6y ago
Yes, but cancer is much less spectacular than missing limbs. Especially when it occurs years later in the next generation.
924.
by rscho 6y ago
The US opiate crisis has its roots deep in US healthcare culture and the blame lies on both sides of the doctor-patient fence. US docs used to prescribe opiates like candies, but US patients expect to feel no pain at all and are extremely a
925.
by rscho 6y ago
Codeine is metabolized to morphine, so it's not really 'in-between'. Plus it's a bad painkiller due to variability in pharmacokinetics. Good practice would be to not use it anymore. Now, there also is stuff such as trama
926.
by rscho 6y ago
Medicine used to be, and in substantial part still is, a disordered heap of hypotheses relying on practical experience and relayed by the stars of the field. There are plenty of strange examples throughout history. For example, we used to b
927.
by rscho 6y ago
Kidneys cannot be repaired. You'll need hemodialysis if you overdose on ibuprofen. I don't know the state of the research on the topic, but I expect locations offering OTC NSAIDs to have issues with acute and chronic kidney injury
928.
by rscho 6y ago
I use J quite a lot. I'm interested in why you're not keen about it? Asking because I'm thinking of adding some form of array language to the statistical/data science Racket framework.
929.
by rscho 6y ago
I know a fair bit of J, but never used K. What do you mean by "more generic"? I'm interested.
930.
by rscho 6y ago
https://i.imgur.com/3QdnQcE.png

More ›