scalablenotions

The chaotic nature of these images, full of local minimas, makes me intuitively think that genetic algorithm approaches to optmizing hyper parameter searching are better suited than regression-based ones.

Auto manufacturers don't have the skills to design or implement good UX, and they don't have the leadership with the skills required to hire the people with the right skills.

Dark Mode doesn't hurt my eyes. Light mode does. "Something something science". Whatever.

As someone who tinkers with these things a lot, I'm happy for the voltage I'm playing with not to be increased. There's plenty of scope to improve the plug design without approaching anything near the fragility of this nVidia adapter. There's no pressing need to increase voltage.

I had this effect laying on a rooftop as a teenager watching a meteor storm. The meteors were streaking in straight lines across/through the sky. The illusion of the "dome" of the sky was broken, and my senses were directly telling me that Earth is hurtling through space, and these things are zipping by it. It was an almost physical sense of mind expansion.

AI Whisperer is totally going to be a job

I think the correct title would technically be "Destructively interfering the hype from the facts of quantum computing"?

Suggest editing to "wimped out" instead of using slang for female genitalia as an insult

NGL I was expecting some kind of database designed for operating on tor

This overstates the difficulty of crowd-sourcing expert consensus. Even if two percent of experts are saying "!X" about the pandemic, then yes, that's still a lot of people and it's easy to find one to counter mainstream opinion "X". However, if you filter inept science journalism, populist narrative, etc., and listen to experts in relevant fields, it's pretty easy very early on to identify the 98%-ish consensus.

The 2% of "everyone" who doesn't know the earth is round, doesn't really impede the reliability of everyone knowing the world is round. Likewise, the "what every expert knows" (maybe not futurist prophesying, but at least about what is true today) is accessible, useful and pretty up-to-date.

If I had to determine whether a planet was hostile, "fired 500 million needles into space" would be pretty freakin high up on my criteria

Internalize that human worth has no connection to ability. Easier said than done. As a thought exercise, think about individuals you care about. How strongly does their productivity correlate with how glad you are they are alive and part of your life? Why do you apply different standards to yourself?

I've thought for a long time that battery swapping is the only way to go for all electric vehicles. A subscription or pay-as-you-go service for battery changing is practical and would make battery replacements costing tens of thousands of dollars a non-thing. I know some batteries are heavy, but if stations can provide petroleum pumps, then they can provide mechanized battery exchange infrastructure, right? Bring on the ISO standards for vehicle batteries!

Does anyone have any reason why this wouldn't work? Perhaps parts being stolen from the rental batteries before they're returned or some such?

My PROdesigner and I had been very clear that the final print should cover 46cm x 27cm, all things considered.

Chronic undersupply is probably the only reason it's not going crazy in 2020. The demand is extremely high, and suppliers are struggling to meet it.

That doesn't match how good researchers find quality information. Critical thinking is a ponderous tool, and can even be counter-productive in isolation. Efficient research techniques should be taught in school, with attention to awareness about personal biases, assessing sources of information and cross-referencing

> humans can't explain or understand how we drive, speak, translate, play chess, etc, so why should we expect to understand how models that do these work?

This implies there's no point in pursuing explainability, but many domains involve inferences where the significant predictors are much easier to abstract at a useful level.

For example, if a DLNN could make suggestions as to how to tune a greenhouse given certain yield objectives, then it's reasonable to pursue heuristic techniques aiming to explain what about the parameters most significantly led to the given suggestions.

I've always thought that people's confidence is pixelation filters was unjustified, and that reversal algorithms are eminently achievable. This goes for pixelation of faces for hiding identities, too. Deep learning or even just genetic algorithms testing lots of possibilities and re-pixelating should be able to infer the original image.

Sure, but it sounds like Gates was taking issue on the basis of interpretability

It's not that odd, when it's reasonable to imagine that an intelligence which creates life would communicate with that life. Genesis and the rest of Judeo-Christian scripture are a compelling candidate for an externally facilitated communiqué, because of their unlikely history, outstanding philology and uncanny foresight

So all these AI SAAS companies are fake?

That's cool, but I want people to be able to independently create content for me to enjoy, which is a bit less of a crock. That will probably takes all their time and they need to pay for ingredients for their pot. Bring on universal basic income I guess.

This seems like someone whose life this doesn't affect wondering why everyone can't just "sit above it all" and approach the topic without emotion, as if it isn't calling their identity and humanity into question and completely emotionally exhausting them

Someone who considers trangender women undeserving of feminist support

I wonder what the consequences would have been of 100 successful enemy missions which were averted?

So far, anyone who lives long enough becomes visually impaired. I hope to be able to still code when I've very old.

Glance at it and then get back to working on the fun things they care about?

> i hate autocorrection. No you don't, or you'd turn it off. I'm sure there's a metaphor in there somewhere.

Dark patterns seem to be becoming standard UX for Google now. I notice so much of it all over their products.

This is one day where you REALLY don't want the elevator to jam on the way to the office.