jebarker

38 ms·

661.
by jebarker 3y ago
Seconded. More than once I've been talked out of pursuing a line of research only to see someone else publish decent results on the same thing a couple of years later.
662.
by jebarker 3y ago
No, it means the same thing in both countries. That UK usage is possibly used by some people but I don't ever recall hearing it in my 30 years there.
663.
by jebarker 3y ago
No, it can't. It can tell you the statistical correlation between your facial features and voters for different parties. That's not the same thing as is suggested here.
664.
by jebarker 3y ago
> Due to its age, Voyager doesn't have any ground simulators Has anyone attempted to build this? Assuming the documentation is available
665.
by jebarker 3y ago
If there's "no going back" doesn't that imply you don't have free will and therefore you can't really choose what you do and don't do?
666.
by jebarker 3y ago
> By the measure that we apply to nearly every other species on earth, humans win. I don't understand - what measure?
667.
by jebarker 3y ago
Presumably they wash though, probably after getting out of bed.
668.
by jebarker 3y ago
What I was meaning was that ChatGPT is not a foundation model since it's been fine-tuned. Although the definition in the link is sufficiently broad you could choose to include it. I can't tell from the OpenAI docs whether it'
669.
by jebarker 3y ago
I don't understand your point. To me GPT-4 is not a foundation model, it's been highly tuned for the chat task. Nobody outside of OpenAI knows what that fine-tuning really involved. So it's impossible to say how much finance
670.
by jebarker 3y ago
How do they know GPT-4 received no specialized financial training?
671.
by jebarker 3y ago
Profitable algorithmic traders are not in the business of moving the field forward. They're in the business of making profits.
672.
by jebarker 3y ago
I'm glad the parrots are helping researchers understand tablet games.
673.
by jebarker 3y ago
Yep, I didn't mean to suggest this was my advice for all people, it's just what has worked for me.
674.
by jebarker 3y ago
I didn't mean to provide advice or make an all-encompassing statement. I just said what has been true for me.
675.
by jebarker 3y ago
I’ve come to believe that time spent on organizational tools and processes is a form of procrastination that’s actually detrimental to focus and productivity. I found much more focus and productivity after I read 4000 weeks and switched to
676.
by jebarker 3y ago
How do you feel about that decision now? I left math after my PhD and often find myself feeling nostalgic about it.
677.
by jebarker 3y ago
It's worth making the feature suggestion via discord or e-mail. I've found the developer to be very responsive.
678.
by jebarker 3y ago
For those in the US trying to do retirement planning I highly recommend trying: https://projectionlab.com/ I first saw it shared on HN and I've been a happy customer for the past year and the ability to compare the imp
679.
by jebarker 3y ago
It's also based on having the money continually invested in the stock market, which becomes more nerve-wracking in retirement I'd imagine since you can't easily weather the storm of a half-decade downturn.
680.
by jebarker 3y ago
I would recommend trying to research what you can about the interview process for specific companies you're interested in. I interview for scientists and engineers for an ML research team but I know that my team (and company's) in
681.
by jebarker 3y ago
I'm also in the position of feeling regret over selling stock to diversify that I would have been much better off holding on to. Even though I understand why logically I shouldn't feel bad, it feels like I made a mistake. I assume
682.
by jebarker 3y ago
Speaking as someone working in the field, I find it amusing how much researchers working on automating human work care about human credit assignment.
683.
by jebarker 3y ago
Doesn't laying down in a fixed position render much of the purpose of the device obsolete?
684.
by jebarker 3y ago
This feels like a "no free lunch" situation. I would imagine that any time saving in approximating the gradients this way would be lost to needing to train for more iterations due to the loss in gradient accuracy. Is that not the
685.
by jebarker 3y ago
> Design your APIs for someone as smart as you, but less tolerant of stupid bulls*t. This is definitely applicable outside of robotics. For example, I work on a large-scale LLM training framework and tend to think this way when thinking
686.
by jebarker 3y ago
> I believe it's fruitless to try to reach that state. Because the harder you try, more difficult it is to reach. Just like meditation. With meditation my experience was that the "trick" is realizing that meditation isn&#x
687.
by jebarker 3y ago
I suppose the idea is that only the US Government has sufficient technology to detect the aliens.
688.
by jebarker 3y ago
That's a big stretch, since 1940s includes 1945. Also, I think the Pentagon was busy with something else in the 1940-1945 time frame.
689.
by jebarker 3y ago
If you can look past the self-aggrandizement and advertising, the article does present useful tools and abstractions for thinking about applying LLMs + NNs to any problem. Wolfram is good at joining dots between ideas from different areas a
690.
by jebarker 3y ago
Likewise in Colorado. I see multiple Rivians everytime I leave the house though. I am in the Boulder Bubble though.

More ›