Bernardo Pagnoncelli comments

Results 6 comments of


                                            Bernardo Pagnoncelli

Document how to add auto-regressive stochastic process

Hello Oscar, so, the whole y thing is a bit mysterious, I will just copy that. But the whole \omega thing is confusing to me. You mention in the comments...

Can a policy graph model MDPs?

Thanks, both for the answers, I agree with all that was said. @joaquimg started hinting at different classes of MDPs, and @odow explained how several classes can be represented at...

PublicationPlots do not check for non-finite data

Thanks for the suggestion. Signed: Unknown person

Add tutorial Example: inventory management

Fair enough, 0.995 as a discount factor can be a bit extreme. Still, sddp.jl handled it pretty well, and the final policy coincided with the theoretical optimal solution for that...

Add tutorial Example: inventory management

Before, always before.

Add tutorial Example: inventory management

Yes, it is the same model. I don't fully agree with what you said. There are two options: 1 - Express the present cost in exact form, using quadratic programming....