There's a lot of people in this thread that don't seem to have caught up with the fact that AMD has worked very hard on their cuda translation layer and for the most part it just works now, you can build cuda projects on amd just fine on modern hardware/software.
Also in this world of accelerator programming, people are writing very specialized codes that target a specific architecture, datatype, and even input shape. So with that in mind, how useful is it to have a generic kernel? You still need to do all the targetted optimization to make it performant.
If you want portablitiy you need a machine learning compiler ala TorchInductor or TinyGrad or OpenXLA.
Nice repeat of history given that AMD started out emphasizing x86 compatibility with Intel's CPUs. It's a good strategy. And open sourcing it means it might be be adapted to other hardware platforms too.
My wife has had an autoimmune disease that’s caused inflammation for years. Absolutely terrible back pain. She had to rent a motorized scooter when we went to Disney World, and could barely walk a few hundred feet before having to stop and rest for years.
Last year she started Zepbound (tirzepatide) and the inflammation went away, we went to Disney World again and she walked happily all day. Absolutely life changing. She didn’t even lose much weight and this was at the lowest dose.
Tirzepatide specifically seems especially beneficial, likewise here. Years of slowly increasing symptoms since I was young, lifelong annoying issues, basically all gone on it. And also not overweight at all, I take below lowest dose and skip most weeks now that it's under control.
Oops! Usually, I just post good stuff I find and the dupe detector catches it and does the thing where it redirects you to the existing post. Apologies!
I'm confused by this post because it seems like his partner got best therapy possible with two surgeries followed by medications without going blind or having other major hormonal issues which can happen after surgery. As correctly stated in this thread prolactinomas aren't a death sentence or even (technically) a brain tumour, and the major risks have been avoided so far. What exactly is being accomplished by a VC deeply researching this case beyond satisfying the valid desire to help your life partner?
Even just fundraising for diseases yields tangible results. Some folks dismissed the viral ice bucket challenge, but it raised over $100 million dollars and funded a lot of research in ALS: https://en.wikipedia.org/wiki/Ice_Bucket_Challenge
Hopefully he accomplishes new knowledge discovery and/or fundraising for research into this disease.
Those efforts were for diseases without treatment. But, prolactinomas already have treatment (surgery plus medication). I always ask myself what is the desired outcome from research effort - in this case it seems to be an attempt to cure/totally eradicate a prolactinoma, which would require hundreds of millions of dollars and custom medication (probably immune checkpoint inhibitors) to do, and this disease already has a treatment available.
A treatment is available but it doesn't seem to work in some cases. The author of the post is motivated to improve that. Honestly, I feel happy when people choose to spend their time trying to solve these types of problems instead of working on things they may not be that interested in or on problems where success doesn't really yield a real benefit to humanity.
Maybe, he spends a lot of time on it and doesn't have success. That seems better to me than spending time on something he doesn't really care about.
It's crazy to me that somewhere along the way we lost physical media as a reference point. Journals and YouTube can be good sources of information, but unless heavily confined to high quality information current AI is not able to judge citation quality to come up with good recommendations. The synthesis of real world medical experience is often collated in medical textbooks and yet AI doesn't cite them nearly as much as it should.
The vast majority of journal articles are not available freely to the public. A second problem is that the business of scientific journals has destroyed itself by massive proliferation of lower quality journals with misleading names, slapdash peer review, and the crisis of quiet retractions.
There are actually a lot of freely available medical articles on PubMed. Agree about the proliferation of lower quality journals and articles necessitating manual restrictions on citations.
I think this summary is reductive, because it ignores the surprisingly dense layers of middle management in hospitals and clinics that are paid more than the medical professionals (and even that ignores external middle managers like PBMs etc).
As my medico wife explained it to me, you have about 1 minute after your heart stops before you start having permanent brain damage. Unless you live inside a hospital your risk of surviving a major heart attack at home is less than 20% at the best of times (and in fact seems to be about 10% or less in USA on average on some quick research).
reply