I'm not super-smart on AI, so I may be missing the boat here, but it would be interesting to feed to a neural network all the medal-winning recipes from the National Homebrew Competition, and see what it spits out. You'd probably have to limit the input to a specific style, or you'd get some really weird beer!
I have that book in my stack to read. Might move it up in the stack now!It's been done (more or less), without AI, in "Designing Great Beers" by Ray Daniels. It's not a recipe book per se, but he crunches a lot of data regarding what winning recipes for each style have in common.
I have that book in my stack to read. Might move it up in the stack now!It's been done (more or less), without AI, in "Designing Great Beers" by Ray Daniels. It's not a recipe book per se, but he crunches a lot of data regarding what winning recipes for each style have in common.
Lately I've been seeing more "projects" and discussion about AI recipe creation. Is that of interest to you? Do you want to brew AI created recipes?
I get it; it's not really about advantage or disadvantage. It's just an interesting exercise. I don't think that an AI could develop the "perfect" beer, any more than I think an AI could develop the "perfect" piece of art. Mostly because there's no such thing!I guess I just don’t see any advantage to using AI for home brewing. There aren’t that many combinations for each style to have the variable inputs. In addition, hop usage is pretty personal (to taste), so don’t see how AI will predict what I like more than I can. Maybe for commercial brewers to hit on an average flavor profile and be novel.
It's a great book and I have created several winning recipes with it over the years.I have that book in my stack to read. Might move it up in the stack now!