xpe

93 ms·

781.
by xpe 2y ago
> Intelligence is closely related to the concept of attractiveness and gravitas. Interesting. I wonder why you make this connection. Do you know? Your choice of definition seems to be what I would call "perception of intelligence&qu
782.
by xpe 2y ago
To continue my earlier comment... I prefer not to call an LLM "intelligent" much less "outrageously intelligent". Why? The main reason is communication clarity -- and by communication I mean the notion of a sender commun
783.
by xpe 2y ago
> I reacted negatively to the idea earlier that agency should be considered an aspect of intelligence. In the hopes of clarifying any misunderstandings of what I mean... I said "agent" in Russell's sense -- a system with g
784.
by xpe 2y ago
Yikes, maybe we can take a step back? I'm not sure where this is coming from, frankly. One anodyne summary of my comment above would be: > Let's think and communicate more clearly regarding intelligence. Stuart Russell offers a
785.
by xpe 2y ago
I regularly pushback against casual uses of the word “intelligence”. First, there is no objective dividing line. It is a matter of degree relative to something else. Any language that suggests otherwise should be refined or ejected from o
786.
by xpe 2y ago
The answer to too much exaggeration about AI from various angles is _not_ more exaggeration. I get the frustration, but exaggerated ranting isn’t intellectually honest nor effective. The AI + software development ecosystem and demographics
787.
by xpe 2y ago
A follow up idea: Don’t forget we modelers are still in a phase that might be characterized as “shifting mindsets”. The capability to model is getting steadily better, but I don’t think the practice of modeling is anywhere close to maturity
788.
by xpe 2y ago
I don’t disagree with this criticism, but it also illustrates a common pattern where modelers tend to attack each other’s work. This criticism is somewhat inevitable because all models are flawed in some way, and reasonable people can disag
789.
by xpe 2y ago
What you say has a lot of truth. And yet, these kinds of models are still useful, provided a broader range of models is also available and socialized. This touches on a key point: just because we can predict something doesn’t mean we should
790.
by xpe 2y ago
Thanks. From the linked page I only see a marketing page that sells the general ideas. It looks like the common pattern where a consultancy sells their expertise, perhaps with a differentiating advantage that their analysis is backed by a q
791.
by xpe 2y ago
A fun read. Everyone has their breaking point… > Naturally, instead of figuring out what library made a breaking change and spending four hours running git bisect, I decided to throw nearly a decade of muscle-memory and workflow refineme
792.
by xpe 2y ago
True. I'll add a few points: - Statistically, more models tends to be better (in the same sense that more information is better) under the assumption that people can tell better from worse. - I like how you talk about _applying_ models
793.
by xpe 2y ago
I get your point of view. But I'm intentionally trying to focus on prediction. There are ways to track ownership of entities and relationships. There are ways to quantify how contracts are awarded, the results, and the impacts. These a
794.
by xpe 2y ago
The solution to a bad model is more models :). I'm joking but serious. "All models are wrong, but some are useful."
795.
by xpe 2y ago
It depends on "good for who" and over what time scale. My dominant model is this: the core of the Trump administration is most interested in remaining in power and building connections. So they do whatever works to garner favor. T
796.
by xpe 2y ago
I'm looking for predictive economic models out in the open. I want to see the underlying assumptions and experiment. I'm not even picky about the kind of model: it could be a relatively simple linear regression model, a systems dy
797.
by xpe 2y ago
1. If an outside-the-system observer cannot detect any flaws in what a RNG outputs, does the outsider have any basis for claiming a lack of randomness? Practically speaking, randomness is a matter of prediction based on what you know. 2. AI
798.
by xpe 2y ago
Ah, the pendulum / cyclic metaphor. It gets trotted out way too often IMO. The pendulum metaphor fails often; here are some examples where it cannot make sensible predictions: species level extinction; major technological shifts; the o
799.
by xpe 2y ago
> On the level where the learning is done and knowledge is represented in these networks there is no evidence anyone really understands how it works. It is hard to assess the comment above. Depending on what you mean, it is incorrect, in
800.
by xpe 2y ago
Nothing to do with it? You certainly don’t mean that. The software running an LLM is causally involved. Perhaps you can explain your point in a different way? Related: would you claim that the physics of neurons has nothing to do with human
801.
by xpe 2y ago
To be clear, the topic of philosophical zombies has to do with consciousness.
802.
by xpe 2y ago
> But the inferences of the resulting neural nets is not an algorithm. Incorrect. The comment above confuses some concepts. Perhaps this will help: consider a PRNG implemented in software. It is an algorithm. The question of the utility
803.
by xpe 2y ago
I for one (along with many thinkers) define intelligence as the extent to which an agent can solve a particular task. I choose the definition to separate it from issues involving consciousness. Both matter of course.
804.
by xpe 2y ago
> The more I look at it, the more I see the concept of general intelligence as a nonsensical one. "nonsensical"? This isn't the right word, is it? The idea of general intelligence is certainly sensical, in the sense that i
805.
by xpe 2y ago
> I would just chuck the idea of general intelligence out the window. It seems to give us nothing anyway. This was the part that I think is exaggerated / overstated.
806.
by xpe 2y ago
Seems to me that the SO moderation level was a design decision. The pros and cons were consciously weighed. As for me, SO found a nice balance, since I dislike the common failure modes of so many other fora.
807.
by xpe 2y ago
I think the claim above simply redefines “quality platform” with the benefit of hindsight.
808.
by xpe 2y ago
An overreaction and/or exaggeration I think. How hard have you looked at the problem? Without a doubt, there are common aspects of many kinds of intelligence.
809.
by xpe 2y ago
Using various metaphors carefully and fluidly is key. No one is sufficient. Not this one, nor any other. I say: go back to basics. One good foundational point is dispelling confusion and conflation around "intelligence". So many p
810.
by xpe 2y ago
Worth noting that frequency space is often considered one dimensional. Adding the phase is what gives the other dimension.

More ›