Dimitrije Marković
Theoretical Neurobiology Meeting
01.03.2020
https://en.wikipedia.org/wiki/Markov_renewal_process#Relation_to_other_stochastic_processes
If Yt≡Xn for t∈[Tn,Tn+1) then the process Yt is called a semi-Markov process
State space S
…
…
Time
For exponentially distributed iid waiting times we have a continuous time Markov chain
A discrete time Markov chain has geometrically distributed waiting times
Shun-Cheng Yu, "Hidden semi-Markov Models: Theory, Algorithms and Applications", Elsevir 2016.
A graphical representation of HSMM
Latent variables
outcomes
f∈{1,2,3}
time step
s∈{A,B}
f∈{1,2,3}
time step
s∈{A,B}
Phase transitions
p(ft∣ft−1)
M Varmazyar, et al., Journal of Industrial Engineering International (2019).
…
Phase transitions
p(ft∣ft−1)
M Varmazyar, et al., Journal of Industrial Engineering International (2019).
Duration distribution
Negative binomial
p(τ)=(τ−1τ+n−2)(1−δ)τ−1δn
…
Phase transitions
p(ft∣ft−1)
State transitions
p(st∣st−1,ft−1)
A
B
A
B
…
History of past outcomes Ot=(o1,…,ot)
Marginal likelihood
Predictive prior
When simulating behaviour γ→∞
For data analysis γ is a free parameter
f∈{1,…,nmax}
⊗
⊗
loss
gain
cue A
cue B
P(ot1)=[31,31,31]
P(ot2)=[ρ1,ρ2,2ρ,2ρ]
Is the experimental setup useful?
Process:
Thanks to:
https://slides.com/dimarkov/active-inference-semi-markov
https://github.com/dimarkov/pybefit
https://journals.plos.org/ploscompbiol/article?rev=2&id=10.1371/journal.pcbi.1006707