RSS

In­fra-Bayesianism

TagLast edit: 10 Mar 2021 5:50 UTC by habryka

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.

See the Infra-Bayesianism Sequence.

In­tro­duc­tion To The In­fra-Bayesi­anism Sequence

Diffractor27 Aug 2020 8:02 UTC
75 points
48 comments13 min readLW link

AXRP Epi­sode 5 - In­fra-Bayesi­anism with Vanessa Kosoy

DanielFilan10 Mar 2021 4:30 UTC
26 points
12 comments35 min readLW link

Ba­sic In­framea­sure Theory

Diffractor27 Aug 2020 8:02 UTC
20 points
10 comments25 min readLW link

Belief Func­tions And De­ci­sion Theory

Diffractor27 Aug 2020 8:00 UTC
12 points
8 comments39 min readLW link

In­framea­sures and Do­main Theory

Diffractor28 Mar 2021 9:19 UTC
26 points
3 comments33 min readLW link

Proofs Sec­tion 1.1 (Ini­tial re­sults to LF-du­al­ity)

Diffractor27 Aug 2020 7:59 UTC
6 points
0 comments20 min readLW link

Proofs Sec­tion 1.2 (Mix­tures, Up­dates, Push­for­wards)

Diffractor27 Aug 2020 7:57 UTC
7 points
0 comments14 min readLW link

Proofs Sec­tion 2.1 (The­o­rem 1, Lem­mas)

Diffractor27 Aug 2020 7:54 UTC
7 points
0 comments36 min readLW link

Proofs Sec­tion 2.2 (Iso­mor­phism to Ex­pec­ta­tions)

Diffractor27 Aug 2020 7:52 UTC
7 points
0 comments46 min readLW link

Proofs Sec­tion 2.3 (Up­dates, De­ci­sion The­ory)

Diffractor27 Aug 2020 7:49 UTC
7 points
0 comments31 min readLW link

Less Ba­sic In­framea­sure Theory

Diffractor16 Dec 2020 3:52 UTC
22 points
1 comment61 min readLW link

The Many Faces of In­fra-Beliefs

Diffractor6 Apr 2021 10:43 UTC
15 points
6 comments63 min readLW link
No comments.