Bayesian reasoning
A method of inference that updates the probability of a hypothesis based on prior beliefs and new evidence; here used in the self-locating argument that one should reason backward about which civilization one is likely in.
As defined in Conversations With Tyler 2021 Retrospective
Where it comes up · 9
- 143 | Julia Galef on Openness, Bias, and Rationality3 claims
- 345 | Adam Elga on Being Rational in a Very Large Universe3 claims
- Conversations With Tyler 2021 Retrospective1 claim
- Conversations with Tyler - Terence Tao1 claim
- Adam Cifu on the Case for Being a Medical Conservative1 claim
- Matjaž Leonardis - Science, Identity and Probability1 claim
- 331 | Solo: Fine-Tuning, God, and the Multiverse1 claim
- AMA | March 20251 claim
and 1 more
Most important claims · 10 of 13Search all claims for “Bayesian reasoning” →
- The Popper-Miller result poses an interesting challenge to using Bayesian reasoning to build an AGI or for related purposes, because it questions whether probabilistic updating captures inductive support.Matjaž Leonardis - Science, Identity and Probability
- The problem of old evidence (raised by Clark Glymour) is especially acute for anthropic reasoning because there is no evidence older than our own existence; it may be unfair to use our existence to update priors since it is a precondition for the conversation rather than a surprising experimental result—an argument Carroll takes seriously but does not ultimately accept.331 | Solo: Fine-Tuning, God, and the Multiverse
- The simulation argument conflates two distinct questions: whether the universe is a simulation in the broad computational sense (almost certainly yes) versus whether we are the product of a specific civilization's experiment (less likely, because civilizations tend to perish before they can create grand simulations, which undermines the Bayesian self-locating reasoning).Conversations With Tyler 2021 Retrospective
- Medicine's reliance on a 5% frequentist significance threshold accepts a high false-positive rate; incorporating Bayesian reasoning that weighs pre-test probability and biological plausibility could reduce false positives, but it raises the danger of who sets the priors, given that study designers already bias studies toward desired answers.Adam Cifu on the Case for Being a Medical Conservative
- Bayesian reasoning is subtle because one may weigh only a hypothesis and a null hypothesis while the true explanation is a third model one has not imagined, so humility about one's own imagination is required; this applies to Bostrom's simulation argument, which Tao calls a good attempt.Conversations with Tyler - Terence Tao
- Confident claims about the existence and near-term discovery of extraterrestrial life often fail proper Bayesian reasoning because expectations get mixed with hope; a good Bayesian should be willing to lower credences in abundant life when searches repeatedly fail, rather than holding fixed beliefs immune to evidence.AMA | March 2025
- David Hume pointed out that rationality cannot tell you what your goals should be; the appeal to extraordinary evidence for extraordinary claims (per Carl Sagan) is a practical application of Bayesian reasoning to the case of reported miracles, where prior improbability is so high that testimony alone is unlikely to overcome it.Steven Pinker Meets Richard Dawkins | On Reason and Rationality
- The claim of Bayesian reasoning is not that it produces 100% accurate beliefs, but that starting from our inevitably flawed and subjective prior beliefs, updating in a roughly Bayesian way will produce more accurate beliefs than not doing so.143 | Julia Galef on Openness, Bias, and Rationality
- Bayes' rule states your confidence in a hypothesis should rise in proportion to the ratio of how likely the observed evidence would be if the hypothesis were true versus if it were false; even without precise numbers, being mindful of this ratio improves reasoning over default human intuition.143 | Julia Galef on Openness, Bias, and Rationality
- The criteria scientists actually use in theory choice—such as treating a constant being close to but not exactly zero as especially bad (fine-tuning), or implicitly coarse-graining over possible parameter values—are far more specific and fine-grained than plain neutral Bayesian probability and cannot be derived from generic priors; they require a cultivated education in physics.345 | Adam Elga on Being Rational in a Very Large Universe