YES(O(1),O(n^1)) 70.53/24.04 YES(O(1),O(n^1)) 70.53/24.04 70.53/24.04 We are left with following problem, upon which TcT provides the 70.53/24.04 certificate YES(O(1),O(n^1)). 70.53/24.04 70.53/24.04 Strict Trs: 70.53/24.04 { g(f(x), y) -> f(h(x, y)) 70.53/24.04 , h(x, y) -> g(x, f(y)) } 70.53/24.04 Obligation: 70.53/24.04 innermost runtime complexity 70.53/24.04 Answer: 70.53/24.04 YES(O(1),O(n^1)) 70.53/24.04 70.53/24.04 The weightgap principle applies (using the following nonconstant 70.53/24.04 growth matrix-interpretation) 70.53/24.04 70.53/24.04 The following argument positions are usable: 70.53/24.04 Uargs(f) = {1} 70.53/24.04 70.53/24.04 TcT has computed the following matrix interpretation satisfying 70.53/24.04 not(EDA) and not(IDA(1)). 70.53/24.04 70.53/24.04 [g](x1, x2) = [1] x1 + [1] x2 + [0] 70.53/24.04 70.53/24.04 [f](x1) = [1] x1 + [0] 70.53/24.04 70.53/24.04 [h](x1, x2) = [1] x1 + [1] x2 + [1] 70.53/24.04 70.53/24.04 The order satisfies the following ordering constraints: 70.53/24.04 70.53/24.04 [g(f(x), y)] = [1] x + [1] y + [0] 70.53/24.04 ? [1] x + [1] y + [1] 70.53/24.04 = [f(h(x, y))] 70.53/24.04 70.53/24.04 [h(x, y)] = [1] x + [1] y + [1] 70.53/24.04 > [1] x + [1] y + [0] 70.53/24.04 = [g(x, f(y))] 70.53/24.04 70.53/24.04 70.53/24.04 Further, it can be verified that all rules not oriented are covered by the weightgap condition. 70.53/24.04 70.53/24.04 We are left with following problem, upon which TcT provides the 70.53/24.04 certificate YES(O(1),O(n^1)). 70.53/24.04 70.53/24.04 Strict Trs: { g(f(x), y) -> f(h(x, y)) } 70.53/24.04 Weak Trs: { h(x, y) -> g(x, f(y)) } 70.53/24.04 Obligation: 70.53/24.04 innermost runtime complexity 70.53/24.04 Answer: 70.53/24.04 YES(O(1),O(n^1)) 70.53/24.04 70.53/24.04 We use the processor 'matrix interpretation of dimension 1' to 70.53/24.04 orient following rules strictly. 70.53/24.04 70.53/24.04 Trs: { g(f(x), y) -> f(h(x, y)) } 70.53/24.04 70.53/24.04 The induced complexity on above rules (modulo remaining rules) is 70.53/24.04 YES(?,O(n^1)) . These rules are moved into the corresponding weak 70.53/24.04 component(s). 70.53/24.04 70.53/24.04 Sub-proof: 70.53/24.04 ---------- 70.53/24.04 The following argument positions are usable: 70.53/24.04 Uargs(f) = {1} 70.53/24.04 70.53/24.04 TcT has computed the following constructor-based matrix 70.53/24.04 interpretation satisfying not(EDA). 70.53/24.04 70.53/24.04 [g](x1, x2) = [2] x1 + [4] 70.53/24.04 70.53/24.04 [f](x1) = [1] x1 + [4] 70.53/24.04 70.53/24.04 [h](x1, x2) = [2] x1 + [4] 70.53/24.04 70.53/24.04 The order satisfies the following ordering constraints: 70.53/24.04 70.53/24.04 [g(f(x), y)] = [2] x + [12] 70.53/24.04 > [2] x + [8] 70.53/24.04 = [f(h(x, y))] 70.53/24.04 70.53/24.04 [h(x, y)] = [2] x + [4] 70.53/24.04 >= [2] x + [4] 70.53/24.04 = [g(x, f(y))] 70.53/24.04 70.53/24.04 70.53/24.04 We return to the main proof. 70.53/24.04 70.53/24.04 We are left with following problem, upon which TcT provides the 70.53/24.04 certificate YES(O(1),O(1)). 70.53/24.04 70.53/24.04 Weak Trs: 70.53/24.04 { g(f(x), y) -> f(h(x, y)) 70.53/24.04 , h(x, y) -> g(x, f(y)) } 70.53/24.04 Obligation: 70.53/24.04 innermost runtime complexity 70.53/24.04 Answer: 70.53/24.04 YES(O(1),O(1)) 70.53/24.04 70.53/24.04 Empty rules are trivially bounded 70.53/24.04 70.53/24.04 Hurray, we answered YES(O(1),O(n^1)) 70.53/24.04 EOF