Hey. How's it going? Transfer learning is how we solve the data quality issue and get Ai truly working in the real world. Translation: Most of data science and Ai today is a complete and utter waste of time and money, (And it's soooo much time and egregious amounts of money that society can't afford to waste). I propose we solve transfer learning with imagination/model based reinforcement learning/agent modeling, (Basically combining all of these in an Agent 57 kinda way).
O and abstraction seems to be necessary for this to happen. That's what multimodal neurons can do.
Yannic Kilcher leading you to literal academic p0rn0gr^phy: • Multimodal Neurons in Artificial Neural Ne...
Multimodal neuron blogpost from OpenAi legends: https://openai.com/blog/multimodal-ne...
CLIP: https://openai.com/blog/clip/
DALE.E: https://openai.com/blog/dall-e/
Doing a patronizing Patreon: / walkerstipe
Doing a PayPal: https://www.paypal.com/paypalme/walke...
Doing a Discord: / discord
Doing a GitHub: https://github.com/walkerstipe (code i share should be available there as well)
F***ing socials…
Twitch: / walkerstipe
Twitter: / walkerstipe
Reddit: / walkerstipe
Linkedin: www.linkedin.com/in/walkerstipe