Tim Quatmann
|
684c239f80
|
Using the revamped optimistic value iteration helper also for the lower bound.
|
6 years ago |
|
Tim Quatmann
|
990d1c9759
|
OVI: seperated implementation from header file. Use a separate helper for computing the upper bounds.
|
6 years ago |
|
Tim Quatmann
|
9ce9ae9eeb
|
OVI: Fixed too early termination.
|
6 years ago |
|
Tim Quatmann
|
2273fda7c9
|
ovi helper: Take the relative/absolute precision and the maximal iteration count as a parameter because otherwise it is not clear whether we should take this information from the NativeEnvironment or the MinMaxEnvironment.
|
6 years ago |
Matthias Volk
|
823ae23180
|
Use updateStatus in more cases
|
6 years ago |
|
Matthias Volk
|
b745b10b77
|
Moved reportStatus() and updateStatusIfNotConverged() to AbstractEquationSolver
|
6 years ago |
|
Matthias Volk
|
d35d31ce94
|
Added SolverStatus::Abort for abort signal handling
|
6 years ago |
|
Tim Quatmann
|
8168b9d125
|
Using OVI as default
|
6 years ago |
|
Tim Quatmann
|
383e2172d4
|
Added OVI for linear equation systems (i.e. DTMC/CTMC)
|
6 years ago |
|
Tim Quatmann
|
e54a035ab9
|
SolverEnvironment: Added the switch `forceExact` to switch to exact solving methods more conveniently.
|
6 years ago |
|
TimQu
|
94fb16e654
|
svi now considers bounds by default
|
8 years ago |
|
TimQu
|
fc43d3f506
|
Added a return type to some lambda expressions as this apparently caused trouble when using gmp numbers
|
8 years ago |
|
TimQu
|
12f8685080
|
Custom Termination Conditions for sound value iteration
|
8 years ago |
|
TimQu
|
8b00f8441e
|
Improved caching for svi
|
8 years ago |
|
TimQu
|
be6d4f9854
|
renamed 'sound power' to 'sound value iteration'
|
8 years ago |
|
TimQu
|
a24de86ce1
|
Avoided duplicated code for sound value iteration
|
8 years ago |
dehnert
|
7150354b9d
|
fixing issue related to vector swapping in (explicit) value iteration and power method
|
8 years ago |
|
TimQu
|
24382630dc
|
removed output of performed iterations to cout
|
8 years ago |
|
TimQu
|
c1ecc22303
|
new multiplyRow method for sound vi
|
8 years ago |
|
TimQu
|
e491dc3813
|
fixed usage of multiplyrow
|
8 years ago |
|
TimQu
|
48945d1199
|
improved multiplyRow method
|
8 years ago |
|
TimQu
|
5ff20b55e1
|
misc compilation issues
|
8 years ago |
|
TimQu
|
f3c843561d
|
integrated new multiplier into native linear equation solver
|
8 years ago |
|
TimQu
|
e76a77abc9
|
improved code for sound power iteration
|
8 years ago |
|
TimQu
|
a6c6118733
|
Renamed 'sound power' to interval iteration and 'quick sound power' to 'sound power'
|
8 years ago |
|
TimQu
|
674a30c154
|
removed some 'experimental' code and introduced the #iteration output again
|
8 years ago |
|
TimQu
|
ca1bcebc71
|
Added an option that enforces updating both bounds in interval iteration
|
8 years ago |
|
TimQu
|
13184aefcf
|
Printing the number of performed iterations (reverted from commit b5399d0596)
|
8 years ago |
|
TimQu
|
b5399d0596
|
Printing the number of performed iterations
|
8 years ago |
|
TimQu
|
5f831d156f
|
forceBounds option for native solver
|
8 years ago |
|
TimQu
|
7eab8589bd
|
Fixed issue in qpower
|
8 years ago |
|
TimQu
|
8c3991fb2f
|
respecting lower/upper bounds from preprocessing in quick sound power method
|
8 years ago |
|
TimQu
|
96845e2669
|
restructured quick power iterations a little
|
8 years ago |
|
TimQu
|
8260455a55
|
Added multiplication of a single matrix row with a vector to the linear equation solver interface
|
8 years ago |
|
TimQu
|
bb0c0bbeb6
|
implemented gauss-seidl multiplications and relative termination for quick power iteration
|
8 years ago |
|
TimQu
|
b55e92bef7
|
Make quick power iteration respect the relevant Values
|
8 years ago |
|
dehnert
|
cd34e3d67e
|
fixed issue in rational search preventing convergence in many cases
|
8 years ago |
|
TimQu
|
4484cea360
|
fixing quick power iteration
|
9 years ago |
|
TimQu
|
3c65a4a10a
|
added a missing assertion
|
9 years ago |
|
TimQu
|
b42aa5f473
|
initial implementation for quick and sound vi for DTMCs
|
9 years ago |
|
TimQu
|
776ce4c8bb
|
Checking requirements of a linear equation solver now depends on whether we want to do multiplication or equation solving. This was necessary to get the correct requirements of a MinMaxSolver that only uses the underlying linear equation solver for multiplication.
|
9 years ago |
|
TimQu
|
e09cb86001
|
making sure that the default linear equation solver is not switched to native if we check e.g. an MDP with sound value iteration
|
9 years ago |
|
TimQu
|
e46f4e154b
|
1. Ensured that when doing policy iteration the underlying solver is at least as precise as the minmax solver.
2. Fix for SolverGuarantees: They are only established if bounds were actually given.
|
9 years ago |
|
TimQu
|
bb63ac6089
|
Linear equation solver + game solvers now respect the environment as well
|
9 years ago |
|
dehnert
|
d90c507431
|
fixed bug in sparse bisimulation quotient extraction related to rewards
|
9 years ago |
|
TimQu
|
5071df5c82
|
made sound value iteration work and respect the correct precision
|
9 years ago |
|
dehnert
|
1f16008b75
|
added proper exception handling to sylvan-based sharpening
|
9 years ago |
|
dehnert
|
d612337ebf
|
added bunch of debug output for aliasing problem
|
9 years ago |
|
dehnert
|
da02237769
|
work towards symbolic rational search
|
9 years ago |
|
dehnert
|
58ca07584d
|
rational search for native linear equation solver and several involved fixes
|
9 years ago |
