Oscar Dowson

Document how to add auto-regressive stochastic process

4

```Julia julia> using SDDP, GLPK julia> model = SDDP.LinearPolicyGraph( stages = 3, lower_bound = 0.0, optimizer = GLPK.Optimizer, ) do subproblem, t # ========================================================================== # Regular SDDP.jl section # #...

@adow031 suggested this

Documentation improvements

1

@bernardokp had these suggestions x-ref #443

Add examples of Stochastic Optimal Control

32

For the INFORMS talk

Ideas to improve multistage stochastic integer programs

9

## Motivation The implementation of SDDiP in SDDP.jl is bad. If you use it, you will probably run into convergence issues! No need to open an issue, I'm aware of...

Stopping rules

7

Without an upper bound, the question of when to terminate SDDP-type algorithms is an open question in the literature. Thus, we need some heuristic for preemptively terminating the algorithm when...

This is probably the correct thing for people to use in practice. At every step of the simulation, we reconverge the policy at the current state. The simulator should accept...

Unfinished TODO item from #332

We need a system of plugins to purge cuts. This may mean rethinking our level-one cut selection. - [ ] Throw away cuts with a high dynamic range - [...

ForwardPass plugins: Part II

1

In https://github.com/odow/SDDP.jl/issues/295, we introduced forwardness plugins. The motivation is for @andrewrosemberg's experiments with different network designs. ## Why do we want different models Transmission models are approximations of reality. If...