/export/starexec/sandbox/solver/bin/starexec_run_complexity /export/starexec/sandbox/benchmark/theBenchmark.koat /export/starexec/sandbox/output/output_files


--------------------------------------------------------------------------------


WORST_CASE(Omega(n^1), O(n^1))
proof of /export/starexec/sandbox/benchmark/theBenchmark.koat
# AProVE Commit ID: 794c25de1cacf0d048858bcd21c9a779e1221865 marcel 20200619 unpublished dirty


The runtime complexity of the given CpxIntTrs could be proven to be BOUNDS(n^1, n^1).

(0) CpxIntTrs
(1) Koat Proof [FINISHED, 25 ms]
(2) BOUNDS(1, n^1)
(3) Loat Proof [FINISHED, 645 ms]
(4) BOUNDS(n^1, INF)


----------------------------------------

(0)
Obligation:
Complexity Int TRS consisting of the following rules:
eval1(A, B, C) -> Com_1(eval2(A, B, C)) :|: A >= B + 1
eval2(A, B, C) -> Com_1(eval1(A, B + 1, C)) :|: A >= C + 1
eval2(A, B, C) -> Com_1(eval1(A, B, C + 1)) :|: A >= C + 1
eval2(A, B, C) -> Com_1(eval1(A - 1, B, C)) :|: C >= A
start(A, B, C) -> Com_1(eval1(A, B, C)) :|: TRUE

The start-symbols are:[start_3]


----------------------------------------

(1) Koat Proof (FINISHED)
YES(?, 6*Ar_0 + 4*Ar_1 + 2*Ar_2 + 2)



Initial complexity problem:

1:	T:

		(Comp: ?, Cost: 1)    eval1(Ar_0, Ar_1, Ar_2) -> Com_1(eval2(Ar_0, Ar_1, Ar_2)) [ Ar_0 >= Ar_1 + 1 ]

		(Comp: ?, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1 + 1, Ar_2)) [ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2 + 1)) [ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0 - 1, Ar_1, Ar_2)) [ Ar_2 >= Ar_0 ]

		(Comp: ?, Cost: 1)    start(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2))

		(Comp: 1, Cost: 0)    koat_start(Ar_0, Ar_1, Ar_2) -> Com_1(start(Ar_0, Ar_1, Ar_2)) [ 0 <= 0 ]

	start location:	koat_start

	leaf cost:	0



Repeatedly propagating knowledge in problem 1 produces the following problem:

2:	T:

		(Comp: ?, Cost: 1)    eval1(Ar_0, Ar_1, Ar_2) -> Com_1(eval2(Ar_0, Ar_1, Ar_2)) [ Ar_0 >= Ar_1 + 1 ]

		(Comp: ?, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1 + 1, Ar_2)) [ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2 + 1)) [ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0 - 1, Ar_1, Ar_2)) [ Ar_2 >= Ar_0 ]

		(Comp: 1, Cost: 1)    start(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2))

		(Comp: 1, Cost: 0)    koat_start(Ar_0, Ar_1, Ar_2) -> Com_1(start(Ar_0, Ar_1, Ar_2)) [ 0 <= 0 ]

	start location:	koat_start

	leaf cost:	0



A polynomial rank function with

	Pol(eval1) = V_1 - V_3

	Pol(eval2) = V_1 - V_3

	Pol(start) = V_1 - V_3

	Pol(koat_start) = V_1 - V_3

orients all transitions weakly and the transition

	eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2 + 1)) [ Ar_0 >= Ar_2 + 1 ]

strictly and produces the following problem:

3:	T:

		(Comp: ?, Cost: 1)              eval1(Ar_0, Ar_1, Ar_2) -> Com_1(eval2(Ar_0, Ar_1, Ar_2)) [ Ar_0 >= Ar_1 + 1 ]

		(Comp: ?, Cost: 1)              eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1 + 1, Ar_2)) [ Ar_0 >= Ar_2 + 1 ]

		(Comp: Ar_0 + Ar_2, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2 + 1)) [ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)              eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0 - 1, Ar_1, Ar_2)) [ Ar_2 >= Ar_0 ]

		(Comp: 1, Cost: 1)              start(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2))

		(Comp: 1, Cost: 0)              koat_start(Ar_0, Ar_1, Ar_2) -> Com_1(start(Ar_0, Ar_1, Ar_2)) [ 0 <= 0 ]

	start location:	koat_start

	leaf cost:	0



Applied AI with 'oct' on problem 3 to obtain the following invariants:

  For symbol eval2: X_1 - X_2 - 1 >= 0





This yielded the following problem:

4:	T:

		(Comp: 1, Cost: 0)              koat_start(Ar_0, Ar_1, Ar_2) -> Com_1(start(Ar_0, Ar_1, Ar_2)) [ 0 <= 0 ]

		(Comp: 1, Cost: 1)              start(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2))

		(Comp: ?, Cost: 1)              eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0 - 1, Ar_1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_2 >= Ar_0 ]

		(Comp: Ar_0 + Ar_2, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2 + 1)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)              eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1 + 1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)              eval1(Ar_0, Ar_1, Ar_2) -> Com_1(eval2(Ar_0, Ar_1, Ar_2)) [ Ar_0 >= Ar_1 + 1 ]

	start location:	koat_start

	leaf cost:	0



A polynomial rank function with

	Pol(koat_start) = V_1 - V_2

	Pol(start) = V_1 - V_2

	Pol(eval1) = V_1 - V_2

	Pol(eval2) = V_1 - V_2

orients all transitions weakly and the transitions

	eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1 + 1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_0 >= Ar_2 + 1 ]

	eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0 - 1, Ar_1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_2 >= Ar_0 ]

strictly and produces the following problem:

5:	T:

		(Comp: 1, Cost: 0)              koat_start(Ar_0, Ar_1, Ar_2) -> Com_1(start(Ar_0, Ar_1, Ar_2)) [ 0 <= 0 ]

		(Comp: 1, Cost: 1)              start(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2))

		(Comp: Ar_0 + Ar_1, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0 - 1, Ar_1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_2 >= Ar_0 ]

		(Comp: Ar_0 + Ar_2, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2 + 1)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_0 >= Ar_2 + 1 ]

		(Comp: Ar_0 + Ar_1, Cost: 1)    eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1 + 1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_0 >= Ar_2 + 1 ]

		(Comp: ?, Cost: 1)              eval1(Ar_0, Ar_1, Ar_2) -> Com_1(eval2(Ar_0, Ar_1, Ar_2)) [ Ar_0 >= Ar_1 + 1 ]

	start location:	koat_start

	leaf cost:	0



Repeatedly propagating knowledge in problem 5 produces the following problem:

6:	T:

		(Comp: 1, Cost: 0)                             koat_start(Ar_0, Ar_1, Ar_2) -> Com_1(start(Ar_0, Ar_1, Ar_2)) [ 0 <= 0 ]

		(Comp: 1, Cost: 1)                             start(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2))

		(Comp: Ar_0 + Ar_1, Cost: 1)                   eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0 - 1, Ar_1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_2 >= Ar_0 ]

		(Comp: Ar_0 + Ar_2, Cost: 1)                   eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1, Ar_2 + 1)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_0 >= Ar_2 + 1 ]

		(Comp: Ar_0 + Ar_1, Cost: 1)                   eval2(Ar_0, Ar_1, Ar_2) -> Com_1(eval1(Ar_0, Ar_1 + 1, Ar_2)) [ Ar_0 - Ar_1 - 1 >= 0 /\ Ar_0 >= Ar_2 + 1 ]

		(Comp: 3*Ar_0 + 2*Ar_1 + Ar_2 + 1, Cost: 1)    eval1(Ar_0, Ar_1, Ar_2) -> Com_1(eval2(Ar_0, Ar_1, Ar_2)) [ Ar_0 >= Ar_1 + 1 ]

	start location:	koat_start

	leaf cost:	0



Complexity upper bound 6*Ar_0 + 4*Ar_1 + 2*Ar_2 + 2



Time: 0.110 sec (SMT: 0.096 sec)


----------------------------------------

(2)
BOUNDS(1, n^1)

----------------------------------------

(3) Loat Proof (FINISHED)


### Pre-processing the ITS problem ###



Initial linear ITS problem

   Start location: start

      0: eval1 -> eval2 : [ A>=1+B ], cost: 1

      1: eval2 -> eval1 : B'=1+B, [ A>=1+C ], cost: 1

      2: eval2 -> eval1 : C'=1+C, [ A>=1+C ], cost: 1

      3: eval2 -> eval1 : A'=-1+A, [ C>=A ], cost: 1

      4: start -> eval1 : [], cost: 1



Checking for constant complexity:

   The following rule is satisfiable with cost >= 1, yielding constant complexity:

      4: start -> eval1 : [], cost: 1



### Simplification by acceleration and chaining ###



Eliminated locations (on tree-shaped paths):

   Start location: start

      5: eval1 -> eval1 : B'=1+B, [ A>=1+B && A>=1+C ], cost: 2

      6: eval1 -> eval1 : C'=1+C, [ A>=1+B && A>=1+C ], cost: 2

      7: eval1 -> eval1 : A'=-1+A, [ A>=1+B && C>=A ], cost: 2

      4: start -> eval1 : [], cost: 1



Accelerating simple loops of location 0.

   Accelerating the following rules:

      5: eval1 -> eval1 : B'=1+B, [ A>=1+B && A>=1+C ], cost: 2

      6: eval1 -> eval1 : C'=1+C, [ A>=1+B && A>=1+C ], cost: 2

      7: eval1 -> eval1 : A'=-1+A, [ A>=1+B && C>=A ], cost: 2



   Accelerated rule 5 with metering function A-B, yielding the new rule 8.

   Accelerated rule 6 with metering function -C+A, yielding the new rule 9.

   Accelerated rule 7 with metering function A-B, yielding the new rule 10.

   Removing the simple loops: 5 6 7.



Accelerated all simple loops using metering functions (where possible):

   Start location: start

      8: eval1 -> eval1 : B'=A, [ A>=1+B && A>=1+C ], cost: 2*A-2*B

      9: eval1 -> eval1 : C'=A, [ A>=1+B && A>=1+C ], cost: -2*C+2*A

     10: eval1 -> eval1 : A'=B, [ A>=1+B && C>=A ], cost: 2*A-2*B

      4: start -> eval1 : [], cost: 1



Chained accelerated rules (with incoming rules):

   Start location: start

      4: start -> eval1 : [], cost: 1

     11: start -> eval1 : B'=A, [ A>=1+B && A>=1+C ], cost: 1+2*A-2*B

     12: start -> eval1 : C'=A, [ A>=1+B && A>=1+C ], cost: 1-2*C+2*A

     13: start -> eval1 : A'=B, [ A>=1+B && C>=A ], cost: 1+2*A-2*B



Removed unreachable locations (and leaf rules with constant cost):

   Start location: start

     11: start -> eval1 : B'=A, [ A>=1+B && A>=1+C ], cost: 1+2*A-2*B

     12: start -> eval1 : C'=A, [ A>=1+B && A>=1+C ], cost: 1-2*C+2*A

     13: start -> eval1 : A'=B, [ A>=1+B && C>=A ], cost: 1+2*A-2*B



### Computing asymptotic complexity ###



Fully simplified ITS problem

   Start location: start

     11: start -> eval1 : B'=A, [ A>=1+B && A>=1+C ], cost: 1+2*A-2*B

     12: start -> eval1 : C'=A, [ A>=1+B && A>=1+C ], cost: 1-2*C+2*A

     13: start -> eval1 : A'=B, [ A>=1+B && C>=A ], cost: 1+2*A-2*B



Computing asymptotic complexity for rule 11

   Solved the limit problem by the following transformations:

      Created initial limit problem:

      -C+A (+/+!), 1+2*A-2*B (+), A-B (+/+!) [not solved]



      removing all constraints (solved by SMT)

      resulting limit problem: [solved]



      applying transformation rule (C) using substitution {C==0,A==1,B==-n}

      resulting limit problem:

      [solved]



   Solution:

      C / 0

      A / 1

      B / -n

   Resulting cost 3+2*n has complexity: Poly(n^1)



Found new complexity Poly(n^1).



Obtained the following overall complexity (w.r.t. the length of the input n):

   Complexity:  Poly(n^1)

   Cpx degree:  1

   Solved cost: 3+2*n

   Rule cost:   1+2*A-2*B

   Rule guard:  [ A>=1+B && A>=1+C ]



WORST_CASE(Omega(n^1),?)


----------------------------------------

(4)
BOUNDS(n^1, INF)