4 ms·
Interesting article, but I think this sentence was unfair to other AI scholars, who also want AI to help society. "While many scholars in the developing field
by Bostonian 5y ago
Interesting article, but I think this sentence was unfair to other AI scholars, who also want AI to help society.
"While many scholars in the developing field of machine learning were focused on improving algorithms, Rudin instead wanted to use AI’s power to help society."
- pjmorris 5y agoA Peter Norvig quote from yesterday's article about his transition to Stanford... "In the past, the interesting questions were around what algorithm is best for doing this optimization. Now that we have a great set of algorithms and tools, the more pressing questions are human-centered: Exactly what do you want to optimize? Whose interests are you serving? Are you being fair to everyone? Is anyone being left out? Is the data you collected inclusive, or is it biased?" [0] https://news.ycombinator.com/item?id=28833933 https://news.ycombinator.com/item?id=28833933
- jhgb 5y agoThat makes the sentence quoted by GP sound like a category error to me. Developing better algorithms and developing more useful models to run on those algorithms are not an "either/or" situation.
- lallysingh 5y agoMany AI researchers do focus on algorithms, no?
- JustFinishedBSG 5y agoPlenty focus on theory.
- tremon 5y agoI think this is meant as a distinction between theoretical and applied science, phrased to a low common denominator (which, as a consequence, causes miscommunication due to not being specific enough).
- qPM9l3XJrF 5y agoMost AI papers I see aren't directly focused on using AI to help society. It's unclear how a small performance increase on ImageNet helps society, for instance.
- jhgb 5y agoI would imagine that it helps by saving computational resources, making the results cheaper to obtain.