However, it is hard to imagine an actual application of the process. If I understand it correctly, the author suggests using a set of micro-models for annotating a dataset which is then used to train another model. The latter model can actually detect Batman in a general environment, ie, can generalize. However, enriching a training dataset by adding adjacent frames depicting Batman from the same movie will likely have limited usefulness when training an actual Batman detection (non-micro!) model. Or do I get the final application wrong?
The key paragraph is the following:
"We are seeing some mutation coming up in some samples that could possibly evade immune responses," said Shahid Jameel, chair of the scientific advisory group of INSACOG and a top Indian virologist. He did not say if the mutations have been seen in the Indian variant or any other strain.
Your point [1] argues that the USSR collapsed due to internal reasons only. I believe that it is an oversimplification.
Cooperation with totalitarian states often prolong their existence. An example that comes to mind is the US subsidizing grain being sold to the USSR and thus preventing starvation, which would have arguably led to a collapse of the communist regime [a].
However, it is also possible to encourage the "feedback loop" that you mention by merely demonstrating the alternatives that are out there. Radio from the other side of the Iron Curtain [b] gave hope to many people in the USSR.
To sum up, thoughtful action from the outside can help bring down totalitarian regimes faster.
This discovery is definitively cool, but not close to the phosphine-on-Venus cool. The present paper is similar to the former finding in the sense that what looked like a natural explanation (a p-wave order parameter) seems to have been ruled out (just like chemical origin of phosphine seems to have been ruled out). Hence, more exotic findings might lurk in this material.
Wiki says that "Hydroxychloroquine is being studied to prevent and treat coronavirus disease 2019 (COVID‑19), but all clinical trials conducted during 2020 found it is ineffective and may cause dangerous side effects."
The recent JAMA only says that Vitamin D does not help with depression. On the other hand, it seems to be useful in preventing respiratory infections, see the following for a review of randomized studies https://doi.org/10.1136/bmj.i6583 .
I fully agree that treatment should be prescribed by a human doctor who can explain and answer questions. However, I would not agree that each of the data inputs to the treatment decision should be generated manually. That is already not the case.
Excellent point. That is what the lengthy prospective study phase as well as periodic auditing after deployment are for.
Please also note that there are several important differences as compared to the automotive industry. First, one could argue that the task at hand is trivial as compared to the self driving car. We are operating in a heavily constrained setting with much better understood data inputs and a hundred-year history of medical professionals trying to classify and systematize them. Moreover, our task is not time-critical. It sometimes takes more than a week for such an image to be reported on.
Full disclosure: I am a cofounder at a startup automating chest X-ray reporting.
It is true that ML algorithms are almost always trained on radiologist labels on the same modality, and thus take in the reader biases. I also agree that some radiologists are better than others as you imply.
As a patient, one does not know who will read their film. IMHO we as an industry should aim not at beating 99.999% of radiologists. We should merely make products which consistently perform not worse than an average radiologist at a particular institution. It is always thrilling to outperform humans with your software, but at the end patient outcomes are what matters. Those are about consistent performance over a long period of time.
Demonstrating this consistent performance is the challenging part, but it is possible to prove it through sufficiently careful and lengthy prospective trials. That’s what we are focusing on, and I would love to see the other players in the industry do the same.
From what I understand, NIST has been running a live "competition" on faces of 9 million people for some time already [0]. They are trying to evaluate racial biases in particular [1]. Was there a reason why you decided to not rely on what NIST was putting out?