3 ms·
The real innovation will come one someone uses a Generative AI to make something, and then use a predictive AI to rate it's accuracy, making it go again until i
by jedberg 4y ago
The real innovation will come one someone uses a Generative AI to make something, and then use a predictive AI to rate it's accuracy, making it go again until it passes the predictive AI.
Basically a form of adversarial training/generation.
- arrow7000 4y agoIsn't this exactly how GANs work already?
- jedberg 4y agoYes. But from I've seen no one has applied it to the latest Generative AIs.
- arrow7000 4y agoMaybe an adversarial approach was used in training these models in the first place?
- sharemywin 4y agoIt was they were' trained using reinforcement learning with human feedback to create the critic.
- jedberg 4y agoI hadn't thought about human feedback being an adversarial system, but I guess that makes sense, since it's basically a classifier saying "you got this wrong".
- dereg 4y agoI’m pretty sure Anthropic’s Claude is doing that. https://scale.com/blog/chatgpt-vs-claude https://scale.com/blog/chatgpt-vs-claude
- ChikkaChiChi 4y agoBilateral "thinking" makes sense, and you can even feed generative AI back into itself for simple error correction. I believe that we'll see the most success/accuracy once you have generative AI compare itself to itself, monitored by a GAN, which then spits out it's answer while retaining some knowledge as to how it came to the conclusion. A tricameral mind.