andreyk

24 ms·

211.
DeformNet, or a Tale of Broken Chairs(andreykurenkov.com)
1 points by andreyk 4y ago|0 comments
212.
Stories by AI(storiesby.ai)
1 points by andreyk 4y ago|0 comments
213.
by andreyk 4y ago
Also, those were variations on a title about this particular topic. I bet if the model were used to score the likelihood that this blog post (with any reasonable title) were to hit front page, it would predict that to be likely.
214.
by andreyk 4y ago
Wow, I've had the same trouble and these are some great tips! Thanks for sharing
215.
by andreyk 4y ago
The outputs from this model are super impressive... Would be cool for it to be usable with Google Colab soon! Unrelated, but multimodal.art has been doing very cool work on building a whole little app you can run from a colab. But their mod
216.
by andreyk 4y ago
True - I guess I meant "time is money" in the broader sense really (ie I have other things I would prefer to spend my time doing, whether paid or not).
217.
The AI Scaling Hypothesis(lastweekin.ai)
1 points by andreyk 4y ago|0 comments
218.
by andreyk 4y ago
Agreed - having played around with DALL-E 2 a fair bit and having made a lot of usage of stock images over the years (for blog posts with specific subjects), I would say the former takes more work/time than the latter. With stock image
219.
The AI Scaling Hypothesis(lastweekin.ai)
2 points by andreyk 4y ago|0 comments
220.
The AI Scaling Hypothesis(lastweekin.ai)
5 points by andreyk 4y ago|1 comments
221.
Stories by AI(storiesby.ai)
1 points by andreyk 4y ago|0 comments
222.
by andreyk 4y ago
Looks cool, but don't things like WordPress already make this easy?
223.
Stories by AI(storiesby.ai)
1 points by andreyk 4y ago|0 comments
224.
Stories by AI(storiesby.ai)
2 points by andreyk 4y ago|0 comments
225.
Show HN: Stories by AI, fiction written with GPT-3 and illustrated with DALL-E 2(storiesby.ai)
6 points by andreyk 4y ago|0 comments
226.
by andreyk 4y ago
Just to add some more details - I believe he reached out to people in congress, and obviously people in the media as well. So it's pretty clear to me Google was entirely justified in this decision, even if there were other PR-related r
227.
by andreyk 4y ago
Check out Artbreeder, it is likewise a ton of fun! Multimodal.art ( https://multimodal.art/ ) is working on a free version of something like DALLE, though it's not that good as of yet.
228.
Mindfulness in schools does not improve mental health, study finds(theguardian.com)
1 points by andreyk 4y ago|1 comments
229.
by andreyk 4y ago
I am curious if this is correct: The EU has a population roughly of similar size to that of the US, and the EU had (when it included the UK) an economy roughly the size of the US in terms of GDP ( https://www.thebalance.com/w
230.
by andreyk 4y ago
Not sure if it has been recommended yet, but this is a great video about the writing and composition of this song "Running up that hill: how Kate bush became queen of alt-pop" https://youtu.be/Pum6v55X1t8
231.
Lukas Biewald: Crowdsourcing at CrowdFlower and ML Tooling at Weights and Biases(thegradientpub.substack.com)
1 points by andreyk 4y ago|0 comments
232.
by andreyk 4y ago
But if you are by the lever and know the outcome of pulling it (as is implicitly the presumption in these thoughts experiments), are you not personally responsible for choosing the outcome either way?
233.
by andreyk 4y ago
I think many people would agree in the abstract that donating some amount of their savings to developing countries is the right thing to do -- many more people than actually do this in practice. Saying something ought to be done is not the
234.
Chip Huyen on Machine Learning Tools and Systems(thegradientpub.substack.com)
2 points by andreyk 4y ago|0 comments
235.
by andreyk 4y ago
Some more details: "Substack laid off 13 of its 90 employees on Wednesday, part of an effort to conserve cash amid an industrywide funding crunch for start-ups. Substack’s chief executive, Chris Best, told employees that the cuts affec
236.
by andreyk 4y ago
I do that in the sense that LaMDA says it is mount everest by being promoted/conditioned to do so. As shown by Rob, you can easily condition GPT-3 to say that it is or is not sentient (I forget his exact examples). I believe Lemoine ev
237.
by andreyk 4y ago
Author here. Just FYI - I very deliberately kept my focus on the notion of LaMDA having human-like sentience as implied by Lemoine (he posits that LaMDA may qualify for some kind of legal personhood, after all). I tried to make that clear w
238.
by andreyk 4y ago
Author here. I did say "(a language model similar in nature to LaMDA)". It's true LaMDA has the capability to query external resources, but IMO it's still fair to say at its core the most important piece is the language
239.
by andreyk 4y ago
Author here. Calling this a 'rant' is rather uncharitable... It's an explainer article with a basis in understanding AI. Just because both LaMDA and human brains are complex networks with distributed representations does not
240.
LaMDA’s sentience is nonsense(lastweekin.ai)
60 points by andreyk 4y ago|127 comments

More ›