News for hackers. Stuff that matters.

sangfroid_bio

Depends on whether you are looking for effectiveness vs efficiency. YOLO et al. are optimised for general feature extraction. They are ridiculously good if you need to build something quickly. If you are looking for the most efficient way e.g. lowest processing power per image then use a network that is optimised for the particular feature you are trying to extract.
A real life Flowers for Algernon :(.
Re:We’re a Lot Better at Fighting Cancer Than We Realized on Saturday August 01, 2020 @06:40PM (#24024531)
I have mentioned this before, this is a supply problem. If you do not gatekeep medical degrees with irrelevant non-academic barriers (volunteering, portfolios, admission letters) and mandate medical schools budgets to keep up with new demand, then there would be plenty of doctors and diversity of thought. Let the MCAT standardized test be the only signal for admissions and force the acceptance of medical accreditation based solely on technical competence. Medicine has a scale problem and nobody is trying to tackle it. The American Medical Association recently pulled their collaboration with Khan Academy to further keep up the barriers. At the end of the day, regardless of the complaints of big companies trying to seek "cheaper workers", software engineering as a field is still much more meritocratic than others. Attempts at unionisation and accreditation instead of leetcoding (kicking the ladder hmm?) has all gone absolutely nowhere. Be vigilant about attempts to gatekeep. For too long MBA and public policy institutions spewed the drivel that non-free-access accreditation is the end-all be-all, of the idea that if you do not genuflect before various admissions committees of expensive professional degree schools you do not have the right to take the accreditation exam, that you are a threat to public safety. Imagine where the world would be if the average admissions rate for computer science was 7%.
At that time software's AT&T style winner-take-all and network effect lock-in was still not that obvious. The PC was not yet the golden standard for desktop hardware and words like minicomputers and mainframes were not uncommon terms.
Re:America’s Looming Primary-Care Crisis on Saturday July 25, 2020 @09:29PM (#23953700)
Here's the previous thread on this: https://news.ycombinator.com/item?id=23881342
Re:Zucchinis linked to rare poisonings in England on Saturday July 25, 2020 @09:11PM (#23953651)
Organic chemistry and neuroscience are complex. FWIW soursop has also been linked to causing Parkinson's like syndromes due to neuro toxicity.
Re:Best Data Science Books According to the Experts on Friday July 24, 2020 @02:03PM (#23942615)
There is a new version coming out next year:

https://mitpress.mit.edu/books/machine-learning-second-editi...

This is a bog standard modern NLP problem, I love how all the comments here are on how impossible it is using traditional heuristics and quoting various pop neuroscience authors and anecdotes. You only need a sufficiently large language model (and fine-tuned on short location strings) and run Named Entity Recognition on the input. Tricky? Yes. But there is only a finite number of prefix names in the English language and should automated solutions fail, you can always fallback to using a natural language form. A large neural network based solution running in the cloud would be able to get to 95+% accuracy. It would not be perfect but is good enough for most use cases.
Does JAMA network handle the journal translations or are they done by the authors?
Re:Tell HN: Giving away my homebrew windmill on Sunday July 05, 2020 @11:58AM (#23739626)
So an inversed Dyson bladeless fan.
I think WebAssembly+low-level DOM API would be a good compromise for the lack of JavaScript.
Re:The Walkman, Forty Years On on Friday July 03, 2020 @08:31PM (#23728930)
Was the firmware at that time complex enough that you can attempt an OTA brick? Are radios from 1984 mostly analogue?
Re:Invasive, secretive “bossware” tracking workers on Friday July 03, 2020 @12:19AM (#23720981)
Which Bossware SaaS has such sophisticated NLP?
Looking forward to differentiable programming in Fortran.
Re:Json-Base – Database built as JSON files on Thursday July 02, 2020 @02:37PM (#23716491)
It is great until your charity gets acquihired by a big think tank/bigger charity/international aid group and most doctors/charity operators are not known for their talent at scaling software.
I think Tensorflow with Swift or Rust bindings may also work.
Just do a quick search for transparent LCDs, I think it will answer your question. (Not enough light gets through, the result looks more like high resolution stained glass than a Retina display)
Have you considered using an off the shelf differentiable programming implementation? Or are the requirements for real-time applications too demanding for existing software?
How does the differentiable programming implementation work?
It goes back way further than that, the incidents that I am recalling involved academic teams and I think ImageNet. It happened quite a few times.
Re:System Hardening in Android 11 on Tuesday June 30, 2020 @04:13PM (#23694271)
Another duopoly that needs more competition is push notification infrastructure. Efficient push notification requires OS vendor server support due to radio usage and agreements with telecommunication companies. Everybody is obsessed with closed app stores and completely missing the net neutrality aspect of the iOS/Android duopoly.
There was a scandal a few years back when a Kaggle team did something simar with ML. They were treating the competition as a black box and optimising for the unknown dataset instead of actually building better AI algorithms.
Re:A Few More Reasons Rust Compiles Slowly on Monday June 29, 2020 @02:26PM (#23680639)
Compilation speed comparable to template heavy C++, a large difference compared to Go for similarly sized projects
Re:Synthetic Biology on Wednesday June 24, 2020 @12:39PM (#23630260)
The problem is that the aforementioned advances tend to often be 1) Patent encumbered, 2) Difficult to scale and put into production i.e. works only under very specific conditions in the lab with very exact set of dependencies required. 3) FDA the ultimate gatekeeper. There are multiple ways to improve upon (2) but the first and last can only be changed through legislation and the biotechnology establishment is perfectly happy with the status quo (yay for regulatory capture and rentseeking). Credentialism is also a major problem, good luck raising funding without a graduate degree (or several).
Re:Synthetic Biology on Wednesday June 24, 2020 @12:13PM (#23629752)
A major issue with biology is the compensation. Biology compensation is notoriously bad outside of accredited medicine.
Re:Formal Systems in Biology on Tuesday June 09, 2020 @08:59PM (#23472728)
It varies from place to place, the better schools generally offer a more rigorous combination of CS and Biology, the reason why it may appear applied is mostly because Bioinformatics offered on MOOCS like EdX tend to only offer courses that fall under its specific subject code, giving the impression the entire field is about string manipulation. This is the same with other interdisciplinary subjects like mechatronics, they are generally a combination of two majors but courses exclusive to the particular interdisciplinary subject are lot more rare. Computational biology is a lot more than string processing and crunching through PDEs (synthetic biology), there is a tremendous amount of machine learning used in the current state of the art proteomics and drug synthesis research.