This topic hasn't been written about much, so the language models still produce fairly straightforward ideas. I may have pulled some inspiration from them, but not much. The great irony of writing about GPT is that it's one area where the best content is going to be human-made
RESEARCH
-
Could 90 Percent of Content Be AI-Driven by 2025?
By
–
#CES: Could 90 Percent of Content Be #AI-Driven by 2025? https://
hollywoodreporter.com/movies/movie-n
ews/ces-ai-sag-aftra-1235290431/
… @THR #CES2023 #innovation #MachineLearning #DataScience #IoT #100DaysofCode #womenwhocode #serverless @CatherineAdenle @Shi4Tech @FmFrancoise @anijov @enilev #NLP #BigData #Analytics #DeepLearning -
GPT-4 Training Data Limitations and Future Improvements
By
–
Yep, huge weakness is all training data is prior to Sep 2021. Really looking forward to GPT-4 to see how current it can be.
-
AI creativity breakthroughs challenge yesterday’s limitations
By
–
It’s important to remember that just two years ago, people were saying AI could never be creative. Just because AI can’t do something today, doesn’t mean won’t be able to soon. The systems are improving faster than anyone expected. Nothing is off limits.
-
Start Known Work Then Test One New Idea
By
–
Depends on how experimental the idea is. Usually you want to start with something you know will work, and then try out one new idea at a time, each time persisting until you understand the results you're seeing fully.
-
Agreement with ChatGPT on Artificial Intelligence Books Selection
By
–
I heartily agree with ChatGPT on four of these. https://
fivebooks.com/best-books/art
ificial-intelligence-gpt-3/
… -
Dexterous Manipulation in Real Life Applications
By
–
dexterous manipulation in real life 👀 https://t.co/zOvBfvoeLi
— Yutaro Yamada (@_yutaroyamada) 8 janvier 2023dexterous manipulation in real life
-
AI Biology Applications: AlphaFold and Beyond Investment Focus
By
–
Biology is where we've seen some of the wildest and most important applications of AI to date. From AlphaFold to [redacted] to [redacted]… Psyched to keep investing in, and start writing about, this intersection with @ElliotHershberg in 2023.
-
Persistence: The Critical Virtue in Machine Learning Engineering
By
–
Perhaps the single most important virtue in ML engineering is persistence. The ML engineering process is one of repeatedly checking & understanding every detail of the system, until it finally goes through a phase transition from "not working at all" to "working shockingly well".
-
GPT-3’s Imitation of Pathological Preferences in Reward Models Discussed
By
–
To be fair, the point Gwern's making (not shown in the quote) isn't just "I don't like it" — it's that GPT-3 is capable of imitating pathological preferences in the reward model.