andreyk

19 ms·

181.
GPT-3 Helps Me Write Short Stories(storiesby.ai)
1 points by andreyk 4y ago|0 comments
182.
Causal Inference: Connecting Data and Reality(thegradient.pub)
1 points by andreyk 4y ago|0 comments
183.
GPT-3 Helps Me Write Short Stories(storiesby.ai)
2 points by andreyk 4y ago|0 comments
184.
Can We Use AI to Communicate with Animals?(lastweekin.ai)
10 points by andreyk 4y ago|5 comments
185.
AI is my co-writer(storiesby.ai)
1 points by andreyk 4y ago|0 comments
186.
by andreyk 4y ago
"“People just get fatigued. They get overwhelmed with the whole dating process,” said Helen Fisher, a biological anthropologist who is a senior research fellow with the Kinsey Institute and chief science adviser to Match.com." I d
187.
by andreyk 4y ago
This is only true for multi modal learning, but yeah, in that case we need text and image pairs. More than likely it's possible to pretrain image and language separately, and then use a vastly smaller number of pairs. But that's h
188.
How to Feel Good(evanjconrad.com)
1 points by andreyk 4y ago|0 comments
189.
by andreyk 4y ago
What are you are describing is pretty much reinforcement learning (or learning with access to a query-able knowledge engine, or active learning, or all of these combined). There is work on a bunch of variations of this, but it's true t
190.
by andreyk 4y ago
A couple of things: * We have seen more than 100,000 "photos" in the sense that photos are just images - if photos are just images, we have a constant feed of "photos" every single moment our eyes are open. Of course, th
191.
by andreyk 4y ago
ah I see, I did not get that it seems. Cool, yeah I think your analysis of the AI's output is quite solid.
192.
by andreyk 4y ago
To your last paragraph, yeah that's kind of the case. We want to use the AI generated as-is more or less to showcase what it produces, so editing those sentences you mention to be more consistent with each other would mess with that. P
193.
by andreyk 4y ago
Got it, my bad! This project was partially inspired by this Show HN "I had some time yesterday so I made a GPT3 podcast to help you sleep" https://news.ycombinator.com/item?id=29428910 So it seemed like the same s
194.
by andreyk 4y ago
wrt "(Although I'm not sure how much human guidance it has)" - at the top it says "Text Formatting: Human-written text is italic, AI-generated text is normal" so that kind of demonstrates the ratio [and the human
195.
A short sci-fi story written with GPT-3 and illustrated with DALL-E 2
125 points by andreyk 4y ago|36 comments
196.
by andreyk 4y ago
Pretty fun post with specific examples of how current tech is not sufficient to take over his job. I have also found when trying to prompt for conceptually specific things (with layout and semantic aspects) it's not that good yet. Will
197.
by andreyk 4y ago
This is somewhat true, but then again it can be argued that the current style is a synthesis of already existing artistic practices from the past (aka 'there is nothing new under the sun'). And if trained on enough data, NNs do ha
198.
by andreyk 4y ago
Seems like a guide for 'being a writer online' aka reach many readers and whatnot. "My life is devoted to helping people write online. In Write of Passage, I teach a proven system that’s built for the 21st century. Our alumni
199.
by andreyk 4y ago
I like this concept - have a one time download link, and a timer to submit solution within an hour or two. Seems like a good balance of fair and realistic (assuming you can't Google for solutions and all).
200.
The AI Scaling Hypothesis(lastweekin.ai)
2 points by andreyk 4y ago|1 comments
201.
In the Era of Large Language Models, What's Next for AI Chatbots?(lastweekin.ai)
3 points by andreyk 4y ago|0 comments
202.
by andreyk 4y ago
Great article! I've personally been pretty self aware with how often I "create doorknobs" in conversations, though I like to think of it in terms of "passing the ball" in soccer or basketball. If you do all the pass
203.
The Future of Speech Recognition: Where Will We Be in 2030?(thegradient.pub)
2 points by andreyk 4y ago|0 comments
204.
by andreyk 4y ago
Can't argue with that!
205.
by andreyk 4y ago
This is using standard prompting I think? It'd be neat to try with the full 'fill in the blank' (where the blank is in the middle of the input) generation technique LLMs can support, might work even better!
206.
by andreyk 4y ago
Well then it will never get boring, will it... Fair enough, it's just been striking to see how many of these sort of posts have been popping up. Like I said, it's still fun and worth a look. Perhaps as a blogger I am extra salty w
207.
What's Next for AI Chatbots in the Era of Large Language Models?(lastweekin.ai)
1 points by andreyk 4y ago|0 comments
208.
by andreyk 4y ago
I wonder how long the novelty of DALL-E will persist. HN seems to upvote anything titled "I did X with DALL-E". This is a fun post, but it's not that interesting or surprising. Still worth a look don't get me wrong, but
209.
by andreyk 4y ago
Gato is definitely cool, but it's worth remembering it was based on supervised learning with data from a bunch of tasks, so data availability is a strong bottleneck with that approach. Still, I'd agree the past year or two have ma
210.
What's Next for AI Chatbots in the Era of Large Language Models?(lastweekin.ai)
2 points by andreyk 4y ago|0 comments

More ›