gradys

7 ms·

181.
by gradys 8y ago
I see a lot of discussion of YIMBY policies like building more market-rate housing. My intuition favors these kinds of policies, but when I've tried to find high quality evidence that they actually achieve objectives like: - Lower hous
182.
by gradys 8y ago
It doesn't need to be deterministic, it just needs to be biased.
183.
by gradys 8y ago
It sounds like you want to be able to generate language with the same basic meaning as the original but without whatever bias was there. I'd recommend starting by trying to detect the presence of the bias first. Language generation is
184.
by gradys 8y ago
How about Epicurianism? https://en.wikipedia.org/wiki/Epicureanism
185.
by gradys 8y ago
Since a number of people who worked on Eve are watching the thread: what's next for you? What do you feel like you learned about human-computer interaction? What are you going to bring to your future work on these kinds of ideas?
186.
by gradys 8y ago
I partially disagree. I studied CS and business as a double major. I work in software now. While it certainly wasn't as important as CS to my career, I'm glad I also did the business major. The technical business skills don't
187.
by gradys 8y ago
I guess it depends on what version of the social contact you think we're all operating with. I'd say the quite minor inconvenience of someone ahead of you in line trying to start a conversation is part of what you sign up for when
188.
by gradys 9y ago
> Try to get them to tell you if they have student loans, a mortgage, a family...anything that indicates a greater immediate financial need. Uh... Might not want to do that actually.
189.
by gradys 9y ago
This depends on water being well-priced in California (i.e. not subsidized, reflective of opportunity costs and externalities). I don't know much about water in CA, but is this the case?
190.
by gradys 9y ago
See https://google.github.io/tacotron/publications/tacotron2/ind... And the HN discussion: https://news.ycombinator.com/item?id=15962543
191.
by gradys 9y ago
Bell Labs? They viewed the development of fundamental technologies as a multi-decade process.
192.
by gradys 9y ago
I'm torn on this. On the one hand, I agree, the EU legislates privacy to a degree that no other political entity does as far as I'm aware, but on the other hand, they often do so with poorly designed policies. As an example, I&#x2
193.
by gradys 9y ago
Tangent: it would be cool to have a browser extension that could notice that this data could be presented visually and offer to show you e.g. the scatter plot.
194.
by gradys 9y ago
Totally agree. It recently won the LA Chinese Film Festival. I'd highly recommend it to anyone who hasn't seen it.
195.
by gradys 9y ago
I'd be interested to see an example of an AI having a fairly coherent conversation. Most of the impressive seeming examples of this based on response ranking. When a message from a human comes in, the system ranks the responses in its
196.
by gradys 9y ago
Maybe not with the techniques they're using now, but if humans can tell it's an ad, then an ad blocker with say an image recognition model can too.
197.
by gradys 9y ago
Here's the associated Google Research Blog post: https://research.googleblog.com/2017/08/harness-power-of-mac...
198.
by gradys 9y ago
I was actually assuming the input representation would just be a video stream, which (combined with audio) is enough for human players, but looking more into it, it's a lot more than a video feed[1]. It feels a little like cheating, bu
199.
by gradys 9y ago
They're discrete with such high cardinality that successful approaches will likely model them assuming they're basically continuous. Neural network layer activations are also discrete after all, but they're often 256+ dimensi
200.
by gradys 9y ago
I think a big component is not really machine learning but more related to how to represent state at any given time, which will necessarily involve a lot of human-tweaking of distilling down what really are the important things that influe
201.
by gradys 9y ago
Why the West Rules for Now - broad coverage of human history, starting with pre-human hominids. Loosely guided by looking for an account of why in 20th century so much power was concentrated in the West compared to the East, but that guidin
202.
by gradys 9y ago
Can people outside of the Ethereum org implement their own languages that target that VM?
203.
by gradys 9y ago
FWIW, I use git inside Dropbox for my personal solo projects. I don't want to have to use commits and git's remote syncing mechanisms to share in-progress code between my desktop at home and my laptop. I want to be able to pick up
204.
by gradys 10y ago
This seems to be more about reducing the risk of reverse engineering the training data of a machine learned system, which could expose sensitive information. Without any special countermeasures, training a model on medical patient data and
205.
by gradys 10y ago
You are missing something vital. The First Amendment guarantees that the government cannot restrict speech, free association, etc. Homeowners associations are not the government. They are private organizations. You generally agree to foll
206.
by gradys 10y ago
At least some of the "toy" problems aren't chosen just for being easy to solve or understand. They're chosen for being qualitatively different than the kinds of problems other neural nets are capable of solving. Sorting,
207.
by gradys 10y ago
I'd extend hypothesis #2. We might not have a sufficiently high-quality theory, but we might also not know how to integrate the benefits of the theory we have with the benefits of data-driven approaches to NLP like those described in t
208.
by gradys 10y ago
Who's to say that neural nets aren't implicitly learning these concepts as they train? In fact, I'd be very surprised if nothing about the networks' internal state corresponded to linguists' models of how language w
209.
by gradys 10y ago
FTA > ... in the coming weeks, they will drive new search algorithms based on deep neural networks—artificial intelligence modeled on the structure of the human brain—executing this AI several orders of magnitude faster than ordinary chi
210.
by gradys 10y ago
From your vantage point, do you see cultural integration (or as you put it, genocide) as anything but inevitable? This article certainly paints a picture that the PRC has every intention of achieving full integration as soon as possible.

More ›