
Will Infra-Bayesianism be built on by 3+ theory or ML papers before 2025?
4
Ṁ90Ṁ271resolved Jan 4
Resolved
NO1H
6H
1D
1W
1M
ALL
"Infra-Bayesianism is a new approach to epistemology / decision theory / reinforcement learning theory, which builds on "imprecise probability" to solve the problem of prior misspecification / grain-of-truth / nonrealizability which plagues Bayesianism and Bayesian reinforcement learning. Infra-Bayesianism also naturally leads to an implementation of UDT, and (more speculatively at this stage) has applications to multi-agent theory, embedded agency and reflection." c.f. this introduction.
Will 3 or more technical papers appear on Arxiv, post-2022, which build on this work before 2025?
This question is managed and resolved by Manifold.
Market context
Get
1,000 to start trading!
🏅 Top traders
| # | Trader | Total profit |
|---|---|---|
| 1 | Ṁ57 | |
| 2 | Ṁ12 | |
| 3 | Ṁ4 |
Sort by:
https://arxiv.org/abs/2405.05673 counts, confirmed with Vanessa Kosoy that other papers citing her work do not meet the criteria for this question.
People are also trading
Related questions
Will I (@Bayesian) be a superforecaster before 2027?
11% chance
Will AIs be widely recognized as having developed a new, innovative, foundational mathematical theory before 2030?
31% chance
Will mainstream science be using primarily Bayesian analysis by 2035?
34% chance
Will AIs be widely recognized as having developed a new, innovative, foundational mathematical theory before 2035?
32% chance
Will we have an AI generated research paper accepted to > 1 top ML conference by 2030?
83% chance
Will we have an AI generated research paper accepted to > 1 top ML conference by 2027?
60% chance
Will we have an AI generated research paper accepted to > 1 top ML conference by 2028?
72% chance
Which 2024 AI paper will have the most citations in 2030?
Will future language models converge on "what Einstein would have thought of Many-Worlds?" before 2036?
46% chance
In 2030, which AI paper will have the most citations?
