fubarobfusco comments on Open thread, 24-30 March 2014 - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (156)
Recently I changed some of my basic opinions about life, in large part because of interaction with LessWrong (mostly along the axes Deism -> Atheism, ethical naturalism -> something else (?)).
It inspired me to try to summarize my most fundamental beliefs. The result is as follows:
1.1. Epistemic truth is to be determined solely by the scientific method / Occam's razor.
1.2. The worldview of mainstream science is mostly correct.
1.3. The many religious / mystical traditions are wrong.
2.1. Consciousness is the result of computing processes in the brain. In particular, if a machine would implement the same computations it would be conscious. However, in general I don't know what consciousness is.
2.2. Identity is not fundamentally meaningful. However, there might be useful "fuzzy" variants of the concept.
3.1. Humans are agents with (approximately) well-defined utility functions.
3.2. The moral value of an action is the expectation value of the utility function of the respective agent.
3.3. I should take actions with as much value as possible. This is the only meaningful interpretation of "should".
4.1. Human utility functions are complex.
4.2. I cannot give anything close to a full description of my utility function, but it seems to involve terminal values such as: beauty, curiosity, humor, kindness, friendship, love, sexuality / romance, pleasure... These values are computed on all sufficiently human agents (but I don't know what "sufficiently human" means). The weights for myself and my friends / loved ones might be higher but I'm not sure.
Less fundamental and less certain are:
5.1. UDT is the correct decision theory.
5.2. Epistemic questions don't make fundamental sense (I realize the apparent contradiction with 1.1 but 1.1 is still a useful approximation and there's also a meta-epistemic level on which UDT itself follows from Occam's razor) as opposed to decision-theoretic questions. Subjective expectations are ill-defined.
5.3. Temark's level IV multiverse is real, or at least as "real" as anything is.
I'm curious to know how many LessWrongers have similar vs different worldviews.
Is this an epistemic truth?
No :) See below.