Object Of Objects @ObjectOfObjects

IMO, THIS. Joined September 2014

Tweets

5K
Followers

2K
Following

234
Likes

1K

Object Of Objects @ObjectOfObjects

11 hours ago

I think influence visualizations in Go shouldn't take into account the turn state. In other words they should use the mean value of the black-to-play and white-to-play situations. Two reasons: 1. Shows when large groups are unsettled 2. Shows what each move achieves

0 0 2 81 0

View Details

Object Of Objects @ObjectOfObjects

2 days ago

@VesselOfSpirit What about the distribution of words? If a word already existed by 2000 but is 10,000x more frequent in 2026, isn't that also suspect?

1 0 1 95 0

View Details

Object Of Objects @ObjectOfObjects

2 days ago

I had an idea about 1-player game strategies. Approximating a game G by a game G+, where the payoff function of G+ is ≥ that of G, gives you both upper and lower bounds on the expected value of G! Upper bound: The E.V. of G+ Lower bound: The E.V. of the G+ policy applied to G.

0 1 2 95 1

View Details

Object Of Objects @ObjectOfObjects

2 days ago

@nostalgebraist @norvid_studies I think thought streams of typical humans also have multiple agents speaking with the same voice as "I."

1 2 4 146 0

View Details

Object Of Objects @ObjectOfObjects

3 days ago

What do people mean by this claim (AI can't create novel ideas)? Is the claim that strings representing new ideas have probability 0 under the model's output distribution?

Zhu Liang @paradite_

7 days ago

i’m really surprised that people don’t see this. It’s mathematically true that llms can’t come up with novel ideas, because the whole point of training is to reduce loss, gain rewards so that the model adhere to rules and ground truth. if you have a model that can come up

480 163 2K 896K 984

1 0 2 198 0

View Details

Object Of Objects @ObjectOfObjects

5 days ago

This reminds me: when performance peaks at a certain age, it doesn't mean that's when age-related decay begins. Growth and decay are happening simultaneously from day one. The peak of the curve is when the growth rate and decay rate are equal.

CognitiveMetrics @CognitiveMetric

5 days ago

A 9-year-old child with an IQ of 130 has about the same intellectual ability as a 100-IQ adult. That is, they answer questions of the same difficulty with the same probability of success. IQ scores are relative and usually age-normed. But because Rasch scores put a person's

36 150 2K 218K 998

0 0 2 256 0

View Details

Object Of Objects @ObjectOfObjects

7 days ago

@badheuristic @khoiiiind The claim I'm responding to is that provability within given bounds is "decidedly decidable," which I interpret as "decidable in practice." A proof of 100 bytes is what I would call very short. Ruling out such a proof costs something like 256¹⁰⁰ checks, unless P = NP.

0 0 1 14 0

View Details

Object Of Objects @ObjectOfObjects

a week ago

I think you could approach the game of Go as having two main topics of study, which are "fighting" and "not fighting," and each of these topics can have a pretty unified treatment.

0 0 1 114 0

View Details

Object Of Objects @ObjectOfObjects

a week ago

@psychiel In principle, how can someone copy your pronunciation without knowledge of the language? They can't reproduce the exact waveform of your voice, so they need to choose the most similar waveform among those they can produce. But the similarity metric depends on the language.

0 0 6 327 0

View Details

Object Of Objects @ObjectOfObjects

a week ago

Isekai protagonist who is granted the power to use magical "save points" but, perplexing, refuses to use them (but is actually using the "save state" feature in the emulator).

0 0 3 84 0

View Details

Object Of Objects @ObjectOfObjects

2 weeks ago

@an_interstice Well, I'm not pushing a single alternative. My claim is that the original scenario is a narrow target. Maybe it looks likelier to some people because of recent history. Your scenario does seem a good deal more likely.

1 0 1 64 0

View Details

Object Of Objects @ObjectOfObjects

2 weeks ago

Prediction: it is not the case that human science and mathematics in the future will consist mainly of trying to make sense of enormous slop results produced by AI.

1 0 2 225 0

View Details

Object Of Objects @ObjectOfObjects

2 weeks ago

@bcubeddd Both claims are backwards. Humanity is innate; adults are more captured by the environment.

0 0 1 142 0

View Details

Object Of Objects @ObjectOfObjects

2 weeks ago

@ApoStructura One reason for this is that it's good to be a monopoly. One big company has more leverage than a bunch of small companies.

0 0 1 56 0

View Details

Object Of Objects @ObjectOfObjects

2 weeks ago

I think I'm really interested in local approximations of a distribution, where the approximation also goes to zero as you move away from the point of approximation. Like clusters, or kernels, or proposal distributions.