So: what’s the deal with Akaike information criterion vs. Bayesian information criterion? "Information theory” and “Bayesianism” are both things with a lot of very devoted adherents and here they appear superficially to give different answers
studyinglogic liked this
khintchine reblogged this from lambdaphagy
vatofbrains liked this
lambdaphagy liked this
neurocybernetics-moved-blog liked this
phenoct liked this
bhishma99 liked this
nostalgebraist reblogged this from raginrayguns and added:
This is all very interesting. Another property you’d want is invariance under general changes of variables, which L2...
thejavaman liked this
raginrayguns reblogged this from lambdaphagy and added:
Okay, I tried to think this through a bit with L2 distance, and I think I’m dropping several levels in HabitRPG as a...
eka-mark liked this
kerapace liked this
Oops, didn’t have a chance to get to this earlier. Others have already chimed in with sensible responses, but here’s...
almostcoralchaos reblogged this from dataandphilosophy
eolianthane reblogged this from nostalgebraist and added:
KL (and AIC) can be put in the MML formalism pretty naturally. KL literally tells you the number of extra bits using a...
shlevy liked this serpent-moon liked this
rangi42 liked this
itthatpoints liked this
znk liked this
automatic-ally liked this
shacklesburst liked this
snarp liked this
youzicha liked this
adzolotl liked this
thespineanditstingle liked this
polyarche liked this
spookyrukie liked this
malpollyon liked this 
bulbous-oar liked this vaniver liked this
vidvilts liked this
dataandphilosophy reblogged this from nostalgebraist and added:
I poked at this a while back, because my textbook introduced both with little information, and what other (unknown...
epistemic-horror reblogged this from nostalgebraist and added:
(IANA real statistician; any or all of this may be wrong)I’m not sure it’s that surprising? The AIC is derived by...
They correspond to different priors. AIC has a bit better underlying framework (from an information theory point of...