3 ms·
Utilitarianism Expressed in Julia
- rdevsrex 5y agoPretty cool. I was thinking about this same thing today and learned that "computational philosophy" is a thing. I've also thought about how truth, not just morals, can be modeled by some sort of "truth space" analogous to a vector space where each area has some certain value and you could sum to decide if a proposition is true or not. I'm curious if machine learning has been applied to such ends.
- platz 5y agono, because you'll never be able to define utility 1) Accurately 2) According to some fixed rule.
- samvega_ 5y agoI always thought deontology (esp. Kantianism) would be a more interesting moral paradigm to implement programmatically. It's all about formalizing what maxims are, and determine whether they are universalizable in a purely formal sense, thus abstracting away from the impossibly difficult task of evaluating the far-reaching consequences of actions. It's less about empirical facts, and more about the internal logic of the concept of willing. I.e., it's about answering the question: what can one rationally will?