P( Doom | AGI ) < P( Doom | no AGI ) #AGI #AGIFirst #ExistentialRisk
RESEARCH
-
Concurrent Research on AI Paper Released Earlier
By
–
actually @ZhongRuiqi was working on this at the same time, and released his paper a few days earlier, I just checked 🙂 https://
arxiv.org/abs/2104.04670 -
Ethics in AI: Deep Dive into Best Practices Newsletter
By
–
Piqued your curiosity? Don't miss out on the complete conversation in the newsletter: https://
louisbouchard.substack.com/p/ethics-in-ai
-a-deep-dive-into-best
… Subscribe to the newsletter for all upcoming episodes with AI experts! -
Responsible AI: Insights from Ethics Research Discussion
By
–
Understanding the ideals of Responsible AI is vital within the realm of modern technology. Here are some insights from my discussion with Auxane Boch, Research Associate and Doctorate Candidate at the Institute for Ethics in AI.
-
Study and Practice: Complementary Approaches to Learning
By
–
"Study isn't a replacement for doing" is true, but we talk a lot less about the inverse: doing isn't a replacement for study. You hear a lot more about the former because it's easier to bias towards inaction. But study gives you the n=many for you to figure out a new n=1.
-
Bayesian Causal Inference: Why You Should Be Excited
By
–
Bayesian Causal inference: why you should be excited https://
bit.ly/3QuWyhe #AI #MachineLearning #DeepLearning #LLMs #DataScience -
PhD Journey: From Academia to Building Towards AI and Educational Content
By
–
My Ph.D. highlights were working on building @towards_AI and making educational videos and podcasts with amazing guests… I didn't quit academia out of discontent, but to pursue something I simply personally enjoyed even more.
-
Pre-trained base LLMs offer more diversity than instruct-tuned models
By
–
Re: 1, yes — see the parenthesis in the second screenshot. Re: 2, you get more diversity from pre-trained base LLMs than the RLHF/instruct-tuned LLMs most people know. You can also compensate by prompting with random dictionary words for inspiration, à la TinyStories.
-
NOT Developing AGI Constitutes Existential Risk
By
–
NOT Developing #AGI Is An Existential Risk. #AGIFirst #MontrealAI #QuebecAI
