🤖 Alexander Lew explains how to automate common-sense tasks using probabilistic scripting. [1]
🧠 Basically, we should encode our domain knowledge probabilistically (declarative), rather than heuristically (imperative).

💭 This reminds me of the process of knowledge elicitation from expert systems.
🔁 Seems the machine learning community might be going full circle once again!

[1]: mrxl.us/2XkwqIR

Emmanuel M. Smith's choices:

Mastodon

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!