sdenton4

32 ms·

721.
by sdenton4 3y ago
"Cos-playing academia" is not bad as a description of lesswrong...
722.
by sdenton4 3y ago
Sometimes this is for pre-reading, which can help focus discussion. It's also helpful for folks who are more readers than listeners to have text to follow along with on the slide.
723.
by sdenton4 3y ago
"Who Framed Roger Rabbit" is a whimsical retelling of the destruction of LA's public transit system on behalf of the car industry, who were considering to get highways built and lock people into car ownership. https:/&#
724.
VideoPoet(blog.research.google)
1 points by sdenton4 3y ago|0 comments
725.
by sdenton4 3y ago
I think no one wants to deal with the combined effort of standardization and anti abuse...
726.
by sdenton4 3y ago
Safer but slightly slower is totally fine. It's even more fine when there's no human attention being wasted on the act of driving itself.
727.
by sdenton4 3y ago
Measuring and normalizing the operations isn't necessarily straightforward - suppose we're in a streaming context (common for encrypting things for the internet), in a real-world environment, where computation time might depend on
728.
by sdenton4 3y ago
I'm not sure why 'perplexity is bad' means 'all LLMs are nearly useless.' Do we use Warwick Davis as the reference height for humans?
729.
by sdenton4 3y ago
What is a 'computation' exactly, and how do you know the duration of the slowest one? I imagine the complexity spirals as you try to answer those two questions...
730.
by sdenton4 3y ago
I have been deeply unimpressed with the ML conference track this last year... There's too many papers, too few reviewers, leading to an insane number of PhD student-reviewers. We've gotten some real nonsense reviews, with some rea
731.
by sdenton4 3y ago
And a notorious mail thief, at that...
732.
by sdenton4 3y ago
For a proper head spin, consider that gravity propagates at the speed of light... And light is not fast enough to escape the black hole. https://www.forbes.com/sites/startswithabang/2019/10/24/this..
733.
by sdenton4 3y ago
Something something reversion to the mean...
734.
by sdenton4 3y ago
Here's an explanation I hit on some time ago: More parameters makes it easier to find solutions with low-energy. Suppose we have a product of two variables z = x * y. And now suppose that the 'correct' product is z=2, and we&
735.
by sdenton4 3y ago
Visual and spatial feedback aren't terribly difficult, though - we already have video-game training gyms, from Atari to GTA. If multimodality and feedback are the main barriers to full AGI, I expect we'll be there soon.
736.
by sdenton4 3y ago
My favorite form of this is in Hedwig and the angry inch... "One day in the late mid-eighties, I was in my early late-twenties. I had just been dismissed from University after delivering a brilliant lecture on the aggressive influence
737.
by sdenton4 3y ago
Yeah, I was surprised that I needed to hit back through an entire game history to get back to the HN discussion... It would be more reasonable to map the 'undo' button to backspace and a visible button, and keep the browser back b
738.
by sdenton4 3y ago
"I can hardly conceptualise that at all!" What, never read Flatland? Truly a classic!
739.
by sdenton4 3y ago
That's not just any character... It's Tim Rogers, purveyor of fine six-hour game reviews. https://www.youtube.com/watch?v=779coR-XPTw
740.
by sdenton4 3y ago
X.com is where you go to save Earth from the alien invasion.
741.
by sdenton4 3y ago
Just roll a d100; it will be about as useful... If you roll over a 75, roll an additional d10 to find out your multiplier score (as in, a 10x programmer).
742.
by sdenton4 3y ago
For music id, you have a static base doing that you're looking to compare to. Bird sing varies by individual, however, so extracting peak frequencies doesn't stand a chance of working. For example, birds might song at a higher or
743.
by sdenton4 3y ago
Pretty cool that evolution has helped us work out consistent and rational solutions to the ethics of gun ownership, abortion, and nuclear proliferation.
744.
by sdenton4 3y ago
So you want a location-sensitive hash, or embedding, and you want it to be noise resistant. The ml approach is to define a family of data augmentations A, and a network N, such that for some augmentation f, we have N(f(x)) ~= N(x). Then we
745.
by sdenton4 3y ago
Hard times for Indian classical music, as well...
746.
by sdenton4 3y ago
Curious how different the performance would be if instead of a 'Hall of Famer' we tell the bot that it is decently-good, but will be deactivated if it can't achieve human-level performance...
747.
by sdenton4 3y ago
Lagos is also high up there... Had some friends basically get robbed by the police on their first night in town. :p
748.
by sdenton4 3y ago
Let us listen to the sweet stories of Sherazad... https://gatherer.wizards.com/pages/card/Discussion.aspx?mult...
749.
by sdenton4 3y ago
When I was a young hacker, around 2002, I went for a couple weeks hike in the Greek mountains north of Delphi with little more than a map and a backpack. I wandered through a lot of tiny villages, slowly disappearing as the youth all moved
750.
by sdenton4 3y ago
I just took my first trip to Japan, and found it amazingly easy to get by. There is actually a lot of English signage, but also just great design of public infrastructure which makes everything kinds easy. (For example, one train transfer t

More ›