Feasible-interior-point algorithms start from a strictly feasible interior point, but infeassible-interior-point algorithms just need to start from an arbitrary positive point, we give a potential reduction algorithm ...Feasible-interior-point algorithms start from a strictly feasible interior point, but infeassible-interior-point algorithms just need to start from an arbitrary positive point, we give a potential reduction algorithm from an infeasible-starting-point for a class of non-monotone linear complementarity problem. Its polynomial complexity is analyzed. After finite iterations the algorithm produces an approximate solution of the problem or shows that there is no feasible optimal solution in a large region. Key words linear complementarity problems - infeasible-starting-point - P-matrix - potential function CLC number O 221 Foundation item: Supported by the National Natural Science Foundation of China (70371032) and the Doctoral Educational Foundation of China of the Ministry of Education (20020486035)Biography: Wang Yan-jin (1976-), male, Ph. D candidate, research direction: optimal theory and method.展开更多
Let Λ = {λ_k} be an infinite increasing sequence of positive integers withλ_k → ∞. Let X = {X(t), t ∈ R^N} be a multi-parameter fractional Brownian motion of index (0 【α 【 1) in R^d . Subject to certain hypot...Let Λ = {λ_k} be an infinite increasing sequence of positive integers withλ_k → ∞. Let X = {X(t), t ∈ R^N} be a multi-parameter fractional Brownian motion of index (0 【α 【 1) in R^d . Subject to certain hypotheses, we prove that if N 【 αd, then there exist positivefinite constants K_1 and K_2 such that, with unit probability, K_1 ≤ φ - p_Λ(X([0,1])~N) ≤ φ -p_Λ(G_rX([0,1])~N)) ≤ K_2 if and only if there exists γ 】 0 such that ∑ from k=1 to ∞ of1/λ_k~γ = ∞, where φ(s) = s^(N/α)(loglog 1/s)^(N/2(α)), φ - p_Λ(E) is the Packing-typemeasure of E,X([0, 1]) N is the image and G_rX([0, 1]~N ) = {(t,X(t)); t ∈ [0,1]~N} is the graph ofX, respectively. We also establish liminf type laws of the iterated logarithm for the sojournmeasure of X.展开更多
文摘Feasible-interior-point algorithms start from a strictly feasible interior point, but infeassible-interior-point algorithms just need to start from an arbitrary positive point, we give a potential reduction algorithm from an infeasible-starting-point for a class of non-monotone linear complementarity problem. Its polynomial complexity is analyzed. After finite iterations the algorithm produces an approximate solution of the problem or shows that there is no feasible optimal solution in a large region. Key words linear complementarity problems - infeasible-starting-point - P-matrix - potential function CLC number O 221 Foundation item: Supported by the National Natural Science Foundation of China (70371032) and the Doctoral Educational Foundation of China of the Ministry of Education (20020486035)Biography: Wang Yan-jin (1976-), male, Ph. D candidate, research direction: optimal theory and method.
基金Supported by the National Natural Science Foundation of China (No.10471148)Sci-tech Innovation Item for Excellent Young and Middle-Aged University Teachers and Major Item of Educational Department of Hubei(No.2003A005)
文摘Let Λ = {λ_k} be an infinite increasing sequence of positive integers withλ_k → ∞. Let X = {X(t), t ∈ R^N} be a multi-parameter fractional Brownian motion of index (0 【α 【 1) in R^d . Subject to certain hypotheses, we prove that if N 【 αd, then there exist positivefinite constants K_1 and K_2 such that, with unit probability, K_1 ≤ φ - p_Λ(X([0,1])~N) ≤ φ -p_Λ(G_rX([0,1])~N)) ≤ K_2 if and only if there exists γ 】 0 such that ∑ from k=1 to ∞ of1/λ_k~γ = ∞, where φ(s) = s^(N/α)(loglog 1/s)^(N/2(α)), φ - p_Λ(E) is the Packing-typemeasure of E,X([0, 1]) N is the image and G_rX([0, 1]~N ) = {(t,X(t)); t ∈ [0,1]~N} is the graph ofX, respectively. We also establish liminf type laws of the iterated logarithm for the sojournmeasure of X.
