(Crossposted from my blog Centerless Set) It’s hard to come up with a workably short function that includes human civilization in its global maximum/minimum. This is a problem because we specify our goals to AI using functions. For any practical function we might use, there’s a set of strange and...
I think the key element that determines how easy a piece of writing is to read is its density of novelty. Novelty can be thought of as the writing equivilant of information. Anything the reader already knows doesn't have to be fully processed, it can just be recalled. Known words,...
Summary: You can prompt your own brain, just as you would GPT-3. Sometimes this trick works surprisingly well for finding inspiration or solving confusing problems. Language Prediction GPT-3 is a language prediction model that takes linguistic inputs and predicts what comes next based on what it learned from training data....
(Crossposted from my blog Centerless Set)
It’s hard to come up with a workably short function that includes human civilization in its global maximum/minimum.
This is a problem because we specify our goals to AI using functions. For any practical function we might use, there’s a set of strange and undesirable worlds that satisfy that function better than our world.
For example, if our function measures the probability that some particular glass is filled with water, the space near the maximum is full of worlds like “take over the galaxy and find the location least likely to be affected by astronomical phenomena, then build a megastructure around the glass designed to keep it full of... (read 731 more words →)
I think the key element that determines how easy a piece of writing is to read is its density of novelty.
Novelty can be thought of as the writing equivilant of information. Anything the reader already knows doesn't have to be fully processed, it can just be recalled. Known words, idioms, and structures don't have to be relearned every time they appear. So only new information (novelty) has to be decoded by the reader.
The higher the density of novelty, the harder a piece of writing is to read.
Shakespeare vs Ordinary Speech
Shakespeare is relatively difficult for modern readers because there are lots of unfamiliar words, linguistic structures, and styles of expression. The reader has... (read 794 more words →)
Summary: You can prompt your own brain, just as you would GPT-3. Sometimes this trick works surprisingly well for finding inspiration or solving confusing problems.
Language Prediction
GPT-3 is a language prediction model that takes linguistic inputs and predicts what comes next based on what it learned from training data. Your brain also responds to prompts, and often does so in a way that (on the surface) resembles GPT-3. Consider the following sequences of words:
A stitch in time saves _____.
The land of the free and the home of the _____.
Harry Potter and the Methods of _____.
If you read these one at a time, you’ll likely find that the last word automatically appears in your... (read 453 more words →)
Thanks for the responses, I'll try to address them individually.
I agree that this doesn't adequately represent our goal, but I think the problem persists even when we add lots of qualifications like "make sure the glass is filled with water for the next five minutes and then lose interest". The maximum of that function might not include a large-scale plan due to limited time, but it could include destroying everything within range except for the facility to prevent interference. It's possible that adding enough qualifications would solve this,... (read 419 more words →)