highfrequency

35 ms·

451.
by highfrequency 6y ago
Yes, remote work has always been a thing. The claim is that COVID will accelerate the trend toward a global and remote workforce. Companies have been forced into large-scale WFH experiments. If the experiments show that productivity is tota
452.
by highfrequency 6y ago
The main potential savings is not from avoiding office leases, it's from hiring engineers for 1/3 of the cost or less from places like Europe, India, and Brazil.
453.
by highfrequency 6y ago
> I had a similar experience, and switching model organisms took my design cycle from 2 weeks to two days, and got huge results out of it. We then put the results back in the original organism and recapitulated the improvement. Then my p
454.
by highfrequency 6y ago
Thanks for sharing your views. Do you see the iteration loop in bio remaining infuriatingly slow, or is this something that improves ever year as tools get better / machines get faster / more biotech microservices arise to remove
455.
by highfrequency 6y ago
> Anybody can go there to learn about, and then do, the kind of cutting-edge bench science—gene editing, synthetic biology—that is generally confined to well-funded academic institutions and private corporations. Cool stuff! Can anyone w
456.
by highfrequency 6y ago
Very cool story! Given that the exciting part of your retirement is learning and making things, I'm curious to hear why you decided not to try starting another business that would allow you to develop your skills in a similar way and a
457.
by highfrequency 6y ago
> single-threaded nonvectorized C wastes on the order of 97% of your computer's computational power Can you elaborate on what this means exactly? For example, is there some reasonable C code that runs 33 times slower than some other
458.
by highfrequency 6y ago
This is interesting! To clarify: in the first two weeks, you don't allow yourself to work on the idea (eg start writing code), but you do allow yourself to research the idea. And you find that the initial research phase usually tur
459.
by highfrequency 6y ago
I disagree that voluntariness and lack of real-world relevance are very central for games. I often find that a task that I put off for a while turns out to be fun and gamelike once I get into flow and find that the problem has interesting s
460.
by highfrequency 6y ago
Is there reason to think that people in the past were more selfless and less economically motivated? I'd wager that throughout history, most great ideas and optimizations came from people who were either seeking profit or other kinds o
461.
by highfrequency 7y ago
This is very informative, thanks for posting! Can you elaborate on the "terrible video experience" of Google Hangouts? As a non power-user of video chat, it works fine for me. And the integration with Google calendar is very conve
462.
Ask HN: How Did Zoom Beat Skype and Google Hangouts?
2 points by highfrequency 7y ago|5 comments
463.
by highfrequency 7y ago
Hang on, any reason why Cloudfare's recent market outperformance will continue ?
464.
by highfrequency 7y ago
Interesting stuff, would love to hear more intuition about why this is difficult. Is it just that we haven't found an economical pathway for metabolizing plastics, or is there a fundamental reason why one should not exist? Hydrocarbons
465.
by highfrequency 7y ago
No, the hard part of trading anything is having edge over other people. Options are not priced assuming there will be a "miraculous recovery" in the next 24 hours. They're priced based on the world's expectation of futur
466.
Transportation, Divergence, and the Industrial Revolution(unenumerated.blogspot.com)
1 points by highfrequency 7y ago|0 comments
467.
by highfrequency 7y ago
Is there no variable charge for network bandwidth?
468.
by highfrequency 7y ago
No, prediction is not the basis of classification. You can build a classifier that looks at a single out of context frame and learns to label the objects in it. It won't be able to predict where those objects move in the next frame--it
469.
by highfrequency 7y ago
Curious if there’s a reason Java/Kotlin wouldn’t have worked? There are a number of fast, statically typed languages with much less developer overhead than C++.
470.
by highfrequency 7y ago
> Given that, you can futz w/ stuff and see if the levy-stable structure of this sort of thing will allow you to fold over different stuff Your writing alternates between being overly complicated and being so casual that it's v
471.
by highfrequency 7y ago
Curious about their definition of “response time” in the graph at the end. They’re quoting ~20 microseconds so I assume this doesn’t involve network hops? Is this just the CPU time it takes a Read State server to do one update?
472.
Record Low Number of Startups in 2019(medium.com)
3 points by highfrequency 7y ago|3 comments
473.
by highfrequency 7y ago
Hi Will, thanks for the reply. One thing I'm curious about: this paper discusses how a neuron that uses a different slope for positive and negative updates will converge to an expectile of the reward distribution. And the behavior is v
474.
by highfrequency 7y ago
In reinforcement learning problems, significant improvement has been found from approximating the distribution of rewards for a particular action, rather than just estimating the mean reward. DeepMind's research suggests that a simil
475.
An algorithm that learns through rewards may show how our brain does too(technologyreview.com)
55 points by highfrequency 7y ago|6 comments
476.
Principles of Effective Research (2004)(michaelnielsen.org)
3 points by highfrequency 7y ago|0 comments