Q
stringlengths
4
3.96k
A
stringlengths
1
3k
Result
stringclasses
4 values
Theorem 3.1 If the system (1.2) satisfies one of the conditions of Theorem 2.1, the origin is a center. Furthermore, the phase diagram near the origin is symmetric w.r.t. a straight line.
Figure 3.1 is the phase diagram of the system (1.2), if its parameters satisfy \( {a}_{1} = {b}_{1} = \) \( {b}_{2} = - {a}_{3} = - {b}_{3} = 1/2,{a}_{2} = 7/4 \) for the condition \( \left( {\mathrm{A}}_{11}\right) \) in Theorem 2.1. In this case, the system takes the form\n\n\[ \dot{x} = - y + 1/2{x}^{2} + 7/{4xy} - ...
Yes
Theorem 4.1 Assume the SGCDRE admits a solution \( {P}_{i}\left( t\right) \in {\mathbb{R}}^{n \times n} \) on \( t \in \left\lbrack {0, T}\right\rbrack \), and the finite horizon LQ optimal control problem (2.1), (2.4) is well-posed. Then, the optimal control in the admissible class \( {\mathcal{U}}_{\mathrm{{ad}}} \) ...
Proof The proof is immediate from Lemma 4.3.
No
Theorem 3.4 Let \( \left\{ {\varepsilon }_{k}\right\} \) be a positive sequence which is convergent to zero. Suppose that \( {z}_{k} \) is a stationary point of the problem (2.11) with \( \varepsilon = {\varepsilon }_{k} \) . If \( \bar{z} \) is an accumulation point of the sequence \( \left\{ {z}_{k}\right\} \) such t...
Proof To prove the above result, we only need to prove that \( {\eta }_{i}^{G * } = 0,{\eta }_{i}^{H * } \geq 0 \) for \( i \in {I}_{00}\left( \bar{z}\right) \) . Obviously, Assumption (*) implies that \( \operatorname{supp}\left( {\bar{\gamma }}^{ * }\right) \cap {I}_{00}\left( \bar{z}\right) = \varnothing \) and \( \...
Yes
Theorem 1.1 Consider the system (1.5) with \( \left| \varepsilon \right| \) small enough. Using the first order Melnikov function, the maximal number of limit cycles which bifurcate from the periodic annulus of the origin of system (1.5) \( {\left. \right| }_{\varepsilon = 0} \) is \( n \) . Moreover, the corresponding...
## 2. Proof of Theorem 1.1\n\nFrom Theorem 2.2 in [9] and Remark 2.3 in [12], we know that the first order Melnikov function \( M\left( h\right) \) of system (1.5) has the following form\n\n\[ M\left( h\right) = {\int }_{\overset{⏜}{{A}_{2}{A}_{1}}}{g}_{1}\left( {x, y}\right) \mathrm{d}x - {f}_{1}\left( {x, y}\right) \...
Yes
Lemma 2.2 For \( h \in \left( {0, + \infty }\right) \) ,\n\n\[ M\left( h\right) = {h\psi }\left( h\right) + \sqrt{h}{\varphi }_{n}\left( \sqrt{h}\right) \]\n\nwhere \( \psi \left( h\right) \) is a polynomial in \( h \) of degree \( \left\lbrack \frac{n - 1}{2}\right\rbrack \) and \( {\varphi }_{n}\left( u\right) \) is ...
Proof From Lemma 2.1, we have\n\n\[ M\left( h\right) = \sqrt{h}{\varphi }_{n}\left( \sqrt{h}\right) + \mathop{\sum }\limits_{{k = 1}}^{m}\left\lbrack {{\alpha }_{k}\left( h\right) {I}_{0,0}^{k}\left( h\right) + {\beta }_{k}\left( h\right) {I}_{1,0}^{k}\left( h\right) + {\gamma }_{k}\left( h\right) {I}_{0,1}^{k}\left( h...
Yes
Corollary 3 When \( m \neq n \), a hypothesis test of level \( \alpha \) for the null hypothesis \( p = q \) has the acceptance region\n\n\[ \n{\operatorname{MMD}}_{b}\left( {\mathcal{F}, X, Y}\right) < \sqrt{\frac{K\left( {m + n}\right) }{mn}}\left( {1 + \sqrt{2\log {\alpha }^{-1}}}\right) .\n\]\n\n(3.1)
Proof When \( p = q \) and \( m \neq n \), we get\n\n\[ \n{\mathrm{E}}_{X, Y}\left\lbrack {{\operatorname{MMD}}_{b}\left( {\mathcal{F}, X, Y}\right) }\right\rbrack = {\mathrm{E}}_{X, Y}\left\lbrack {\mathop{\sup }\limits_{{f \in \mathcal{F}}}\left\lbrack {\frac{1}{m}\mathop{\sum }\limits_{{i = 1}}^{m}f\left( {x}_{i}\ri...
Yes
Lemma 2.1 Suppose (V) and (K) hold. Let \( k \mathrel{\text{:=}} \frac{\alpha }{p - 1} \) . Then there exist \( \delta > 0 \) and \( C \geq 1 \) such that for all \( {u}_{0} \in {L}_{k}^{\infty } \) satisfying \( {\begin{Vmatrix}{u}_{0}\end{Vmatrix}}_{\infty, k} \leq \delta \), the corresponding solution \( u \) of (1....
Proof Without loss of generality, assume \( u \geq 0 \) . Let \( U\left( x\right) \mathrel{\text{:=}} \frac{1}{{\left( A + {\left| x\right| }^{2}\right) }^{\frac{k}{2}}} \) for \( A > 0 \), then\n\n\[ {\Delta U} = \mathop{\sum }\limits_{{i = 1}}^{n}\frac{{\partial }^{2}U}{\partial {x}_{i}^{2}} = \mathop{\sum }\limits_{...
Yes
Corollary 2.1 Under the assumptions of Lemma 2.1, if \( {u}_{0} \in {L}_{k,0}^{\infty } \), then \( \mathop{\lim }\limits_{{t \rightarrow \infty }}\parallel u\left( t\right) {\parallel }_{\infty, k} = \) 0.
Proof (2.2) also implies that \( u \leq \frac{{C}^{\prime }\delta }{1 + {\left| x\right| }^{k}} \) . Then for all \( x \) and \( t > 0 \), taking \( 0 < \delta < \) \( {\left( \frac{{a}_{1}}{2{C}^{\prime \prime }}\right) }^{\frac{1}{p - 1}} \), we have\n\n\[ \n{u}_{t} - {\Delta u} + \frac{1}{2}V\left( x\right) u = K\le...
Yes
Lemma 4.1 Let \( {u}_{0} \in {L}^{\infty } \cap {H}^{1} \) . If \( T\left( {u}_{0}\right) = \infty \), then for each \( k \geq 0 \), the \( \omega \) -limit set \( {\omega }_{k}\left( {u}_{0}\right) \) consists of equilibria.
Proof Assume \( \widetilde{v} \in {\omega }_{k}\left( {u}_{0}\right) \), then there exists \( {t}_{j} \rightarrow \infty \) such that \( u\left( {t}_{j}\right) \rightarrow \widetilde{v} \) in \( {L}_{k}^{\infty } \) . Fixing \( t > 0 \) and by the continuity of semigroup, we have\n\n\[ u\left( {t + {t}_{j}}\right) = S\...
Yes
Lemma 5.2(Uniform a priori estimate for global solutions) If \( T\left( {u}_{0}\right) = \infty \), then\n\n\[ \mathop{\sup }\limits_{{t \geq 0}}\parallel u\left( t\right) {\parallel }_{\infty } \leq C\left( {{\begin{Vmatrix}{u}_{0}\end{Vmatrix}}_{\infty } + {\begin{Vmatrix}{u}_{0}\end{Vmatrix}}_{{H}^{1}}}\right) \]\n\...
Proof By the standard local theory, there exists \( \tau = \tau \left( {\begin{Vmatrix}{u}_{0}\end{Vmatrix}}_{\infty }\right) > 0 \) such that\n\n\[ {\begin{Vmatrix}u\left( t;{u}_{0}\right) \end{Vmatrix}}_{\infty } \leq {\begin{Vmatrix}{u}_{0}\end{Vmatrix}}_{\infty } + 1,\;0 \leq t \leq \tau . \]\n\n(5.4)\n\nArguing by...
No
Lemma 3.1 Assume that the conditions \( \mathrm{H}\left( \xi \right) ,\mathrm{H}\left( \beta \right) ,\left( {\mathrm{H}}_{1}\right) \) and \( \left( {\mathrm{H}}_{2}\right) \) hold. Then \( u = 0 \) is a local minimum of \( \mathcal{J} \) and \( {\mathcal{J}}_{ \pm } \) for any \( \lambda \leq 0 \) .
Proof From the conditions \( \left( {\mathrm{H}}_{1}\right) \) and \( \left( {\mathrm{H}}_{2}\right) \), for given \( \epsilon > 0 \), there exists \( {A}_{1} > 0 \) such that\n\n\[ \left| {F\left( {x, u}\right) }\right| \leq \frac{p\left( x\right) + \epsilon }{2}{\left| u\right| }^{2} + {A}_{1}{\left| u\right| }^{\gam...
Yes
Lemma 3.3 Assume that \( \lambda \leq 0 \), and the conditions \( H\left( \xi \right), H\left( \beta \right) ,\left( {\mathrm{H}}_{1}\right) ,\left( {\mathrm{H}}_{2}\right) ,\left( {\mathrm{H}}_{3}\right) \) with \( l = + \infty \) and \( \left( {\mathrm{H}}_{4}\right) \) hold. If \( {u}_{ \pm } \) are the isolated non...
Proof Let\n\n\[{\mathcal{J}}_{s}\left( u\right) = s{\mathcal{J}}_{ + }\left( u\right) + \left( {1 - s}\right) \mathcal{J}\left( u\right), s \in \left\lbrack {0,1}\right\rbrack ,\]\n\nthen \( {\mathcal{J}}_{s} \) satisfies the condition \( {\left( \mathrm{C}\right) }_{c} \) by Lemma 3.2, and \( {u}_{ + } \) is a critica...
Yes
Lemma 3.6 Under the conditions \( \mathrm{H}\left( \xi \right) ,\mathrm{H}\left( \beta \right) ,\left( {\mathrm{H}}_{1}\right) ,\left( {\mathrm{H}}_{2}\right) ,\left( {\mathrm{H}}_{3}\right) \) with \( l = + \infty \) and \( \left( {\mathrm{H}}_{4}\right) \) , the functionals \( \mathcal{J} \) and \( {\mathcal{J}}_{ \p...
Proof We only sketchily give the proof of \( {\mathcal{J}}_{ + } \), the cases of \( \mathcal{J} \) and \( {\mathcal{J}}_{ - } \) are similar. This proof is essentially equal to our the previous section of the proof of Lemma 3.2 and the last section of the proof of Proposition 3 of [9]. Hence, we omit it here.
No
Theorem 3.7 If the conditions \( \mathrm{H}\left( \xi \right) ,\mathrm{H}\left( \beta \right) ,\left( {\mathrm{H}}_{1}\right) \) and \( \left( {\mathrm{H}}_{2}\right) \) hold, then there exists \( {\lambda }^{ * } > 0 \) such that for all \( 0 < \lambda < {\lambda }^{ * } \) we can find \( \rho > 0 \) for which we have...
Proof By the conditions \( \left( {\mathrm{H}}_{1}\right) \) and \( \left( {\mathrm{H}}_{2}\right) \), given \( \epsilon > 0 \), there exists \( {A}_{3} > 0 \) such that\n\n\[ \n{f}_{ + }\left( {x, u}\right) \leq \frac{p\left( x\right) + \epsilon }{2}{\left( {u}^{ + }\right) }^{2} + {A}_{3}{\left( {u}^{ + }\right) }^{\...
Yes
Theorem 4.1 Equations (4.9),(4.11) indeed admit a unique solution \( {\phi }^{ * }\left( t\right) = \left( {{\phi }_{1}^{ * }\left( t\right) ,{\phi }_{2}^{ * }\left( t\right) }\right. \) , \( \left. {{\phi }_{3}^{ * }\left( t\right) ,{\phi }_{4}^{ * }\left( t\right) ,{\phi }_{5}^{ * }\left( t\right) }\right) ,{u}^{ * }...
Proof It is easy to verify that \( {W}_{1}\left( {\beta }_{1}\right) ,{W}_{2}\left( {\beta }_{2}\right) \) defined by (4.12) is an increasing function of the variable \( {\beta }_{1},{\beta }_{2} \), respectively. Moreover, \( {W}_{1}\left( 0\right) = - \left( {{r}_{1} - {r}_{0} + \left( {\lambda + {\lambda }_{1}}\righ...
Yes
Theorem 4.2 Let \( W\left( {\phi }_{6}\right) = \frac{\gamma {h}^{P}}{{\vartheta }_{6}}{\phi }_{6}\ln {\phi }_{6} + {h}^{P}{\phi }_{6} - \frac{\delta }{\varsigma } \), then \( W\left( {\phi }_{6}\right) \) has a unique positive root \( {\phi }_{6} \) .
Proof Since \( {W}^{\prime }\left( {\phi }_{6}\right) = \frac{\gamma {h}^{P}}{{\vartheta }_{6}}\left( {\ln {\phi }_{6} + 1}\right) + {h}^{P} \), it is easy to verify \( W\left( {\phi }_{6}\right) \) is a decreasing function on \( \left( {0,{\mathrm{e}}^{-\frac{\gamma + {\vartheta }_{6}}{\gamma }}}\right) \) and increas...
Yes
Theorem 4.3 If there exists a function and a control policy \( \left( {{\alpha }^{ * }\left( t\right) ,{\beta }_{1}^{ * }\left( t\right) ,{\beta }_{2}^{ * }\left( t\right) ,{\beta }_{3}^{ * }\left( t\right) }\right) \) , which satisfy the HJB equation (4.1), then \( M\left( {t, x, l, z}\right) \) is the corresponding v...
Proof From [11], the above theorem will hold if \( \left( {{\alpha }^{ * }\left( t\right) ,{\beta }_{1}^{ * }\left( t\right) ,{\beta }_{2}^{ * }\left( t\right) ,{\beta }_{2}^{ * }\left( t\right) }\right) \) and the corresponding candidate value function \( M\left( {t, x, l, z}\right) \) has the following three properti...
No
Theorem 3.1 Consider Eq.(1.1), \( a\left( t\right), b\left( t\right) \) and \( c\left( t\right) \) are all \( \omega \) -periodic continuous functions on \( \mathbb{R} \) . Suppose that the following conditions hold:\n\n\( \left( {\mathrm{H}}_{1}\right) a\left( t\right) > 0 \)\n\n\( \left( {\mathrm{H}}_{2}\right) b\lef...
Proof 1) We prove that Eq.(1.1) has two positive \( \omega \) -periodic continuous solutions \( {\gamma }_{1}\left( t\right) \) and \( {\gamma }_{2}\left( t\right) \).\n\nDefine a set as follows:\n\n\[ B = \left\{ {\varphi \left( t\right) \in C\left( {\mathbb{R},\mathbb{R}}\right) \left| {\;\varphi \left( {t + \omega }...
Yes
Theorem 3.2 Under the conditions of Theorem 3.1, Eq.(1.1) has exactly two \( \omega \) -periodic continuous solutions: \( {\gamma }_{1}\left( t\right) \) and \( {\gamma }_{2}\left( t\right) \) .
Proof The proof of the existence of \( {\gamma }_{1}\left( t\right) \) and \( {\gamma }_{2}\left( t\right) \) is given in Theorem 3.1. Now, we prove that Eq.(1.1) has exactly two \( \omega \) -periodic continuous solutions: \( {\gamma }_{1}\left( t\right) \) and \( {\gamma }_{2}\left( t\right) \) . We know if \( x\left...
Yes
Corollary 3.1 Consider the following Bernoulli's equation:\n\n\[ \frac{\mathrm{d}x}{\mathrm{\;d}t} = a\left( t\right) {x}^{2} + b\left( t\right) x \] \n\n\( \left( {3.47}\right) \) \n\nand \( a\left( t\right), b\left( t\right) \) are both \( \omega \) -periodic continuous functions on \( \mathbb{R} \) . Suppose that th...
1) One \( \omega \) -periodic continuous solution is \( {\gamma }_{1}\left( t\right) = 0 \), and \( {\gamma }_{1}\left( t\right) \) is attractive if given initial value on \( {D}_{1} = \left\{ {x\left( {t}_{0}\right) \mid x\left( {t}_{0}\right) < \frac{1}{\zeta \left( {t}_{0}\right) }}\right\} \), and unstable if given...
Yes
Theorem 3.3 Consider Eq.(1.1), \( a\left( t\right), b\left( t\right) \) and \( c\left( t\right) \) are all \( \omega \) -periodic continuous functions on \( \mathbb{R} \) . Suppose that the following conditions hold:\n\n\( \left( {\mathrm{H}}_{7}\right) a\left( t\right) < 0 \)\n\n\( \left( {\mathrm{H}}_{8}\right) b\lef...
Proof Consider the following equation:\n\n\[ \frac{\mathrm{d}x}{\mathrm{\;d}t} = \widehat{a}\left( t\right) {x}^{2} + \widehat{b}\left( t\right) x + \widehat{c}\left( t\right) ,\]\n\n(3.48)\n\nhere \( \widehat{a}\left( t\right) = - a\left( t\right) > 0,\widehat{b}\left( t\right) = - b\left( t\right) < 0,\widehat{c}\lef...
Yes
Theorem 3.4 Under the conditions of Theorem 3.3, Eq.(1.1) has exactly two \( \omega \) -periodic continuous solutions: \( {\gamma }_{1}\left( t\right) \) and \( {\gamma }_{2}\left( t\right) \) .
Proof Theorem 3.4 also follows by applying Theorem 3.2 to the equation (3.48).
No
Corollary 3.2 Consider Bernoulli’s Eq.(3.47), and \( a\left( t\right), b\left( t\right) \) are both \( \omega \) -periodic continuous functions on \( \mathbb{R} \) . Suppose that the following conditions hold:\n\n\( \left( {\mathrm{H}}_{11}\right) a\left( t\right) < 0 \)\n\n\( \left( {\mathrm{H}}_{12}\right) b\left( t\...
1) One \( \omega \) -periodic continuous solution is \( {\gamma }_{1}\left( t\right) = 0 \), and \( {\gamma }_{1}\left( t\right) \) is unstable on \( \mathbb{R} \) ;\n\n2) Another \( \omega \) -periodic continuous solution is\n\n\[ \n{\gamma }_{2}\left( t\right) = - \frac{1}{{\int }_{-\infty }^{t}{\mathrm{e}}^{-{\int }...
Yes
Consider the following equation:\n\n\\[ \n\\frac{\\mathrm{d}x}{\\mathrm{\\;d}t} = \\left( {2 + \\sin t}\\right) {x}^{2} + \\left( {\\sin t - 8}\\right) x + 2 - \\cos t. \n\\]\n\n(4.1)
It is easy to calculate that \\( {a}_{M} = 3,{a}_{L} = 1,{b}_{M} = - 7,{b}_{L} = - 9,{c}_{M} = 3,{c}_{L} = 1 \\), and\n\n\\[ \n{b}_{M}^{2} - 4{a}_{M}{c}_{M} = {13} > 0. \n\\]\n\nClearly, the conditions \\( \\left( {\\mathrm{H}}_{1}\\right) - \\left( {\\mathrm{H}}_{4}\\right) \\) of Theorem 3.1 are satisfied. It follows...
Yes
Consider the following equation:\n\n\[ \frac{\mathrm{d}x}{\mathrm{\;d}t} = \left( {-2 + \sin t}\right) {x}^{2} + \left( {\sin t + 8}\right) x - 2 + \cos t. \]
It is easy to calculate that \( {a}_{M} = - 1,{a}_{L} = - 3,{b}_{M} = 9,{b}_{L} = 7,{c}_{M} = - 1,{c}_{L} = - 3 \), and\n\n\[ {b}_{L}^{2} - 4{a}_{L}{c}_{L} = {13} > 0. \]\n\nClearly, the conditions \( \left( {\mathrm{H}}_{7}\right) - \left( {\mathrm{H}}_{10}\right) \) of Theorem 3.3 are satisfied. It follows from Theor...
Yes
Lemma 2.4 If \( \psi \notin {\delta }_{2} \), then \( {\lambda }_{\varphi ,\omega }^{ \circ } \) contains \( {\ell }_{1} \) .
Proof For any \( \varepsilon > 0 \), by \( \psi \notin {\delta }_{2} \) there is a sequence \( \left\{ {u}_{i}\right\} \) of positive numbers such that \( \varphi \left( {{2}^{2}{u}_{1}}\right) \omega \left( 1\right) < \frac{\varepsilon }{2}, \n\n\[ \n\varphi \left( {{2}^{i + 1}{u}_{i}}\right) \leq \left( {{2}^{i + 1} ...
Yes
Lemma 2.5 If \( \omega \) is not regular, then \( {\lambda }_{\varphi ,\omega }^{ \circ } \) contains \( {\ell }_{\infty }^{n} \) uniformly.
Proof Since \( \omega \) is not regular, we see that for any \( n \) and any \( \varepsilon \in \left( {0,1}\right) \), there exists a nature number \( \alpha > 1 \) such that\n\n\[\n\mathop{\sum }\limits_{{t = 1}}^{{n\alpha }}\omega \left( t\right) \leq \left( {1 + \varepsilon }\right) \mathop{\sum }\limits_{{t = 1}}^...
Yes
Theorem 4.1 Orlicz-Lorentz sequence space \( {\lambda }_{\varphi ,\omega }^{ \circ } \) is uniformly non-square if and only if\n\n(a) \( \omega \) is regular,\n\n(b) \( \varphi ,\psi \in {\delta }_{2} \) .
Proof (Necessity) By Lemmas 2.3-2.4, we can see \( \varphi \in {\delta }_{2} \) and \( \psi \in {\delta }_{2} \) . Suppose \( \omega \) is not regular, then we can find an infinite sequence \( \left\{ {t}_{n}\right\} \subset \mathbb{N} \) such that\n\n\[ \frac{S\left( {2{t}_{n}}\right) }{S\left( {t}_{n}\right) } \right...
Yes
Proposition 3.2 Let \( {H}_{0} \) be a Hamiltonian function of the form \( {H}_{0} = N\left( p\right) + \epsilon \widetilde{H}\left( {p, q}\right) \) with \( N,\widetilde{H} \in {\mathfrak{A}}_{{p}^{ * },\varrho } \) . Denote\n\n\[ \n{\overrightarrow{\lambda }}_{0} = \left( {{\partial }_{{p}_{1}}N\left( {p}^{ * }\right...
Proof Step 1 Let \( \lambda = \nabla N\left( {p}^{ * }\right) \) be the unperturbed frequency. By the classic KAM theorem, there exists a symplectic differmorphism \( \Phi : {B}_{{p}_{ * },\frac{1}{2}\rho } \rightarrow {B}_{{p}_{ * },\frac{1}{2}\rho } \) such that\n\n\[ \nH \circ \Phi = z + \lambda \cdot \left( {p - {p...
Yes
Theorem 2.1 Under the validity of the first-order condition in (1.2), and for intermediate sequences \( k \), i.e., sequences of integer values \( k = k\left( n\right) ,1 \leq k < n \), such that\n\n\[ k = k\left( n\right) \rightarrow \infty ,\;k/n \rightarrow 0,\;\text{ as }\;n \rightarrow \infty ,\]\n\nthe \( {N}_{\a...
Proof Denote \( {K}_{\alpha }\left( \frac{i}{k + 1}\right) \mathrel{\text{:=}} {K}_{\alpha } \) and by Lemma 2.1, we have\n\n\[ {U}_{i} \approx \gamma {E}_{i} + A\left( {n/k}\right) {\left( \frac{i}{k + 1}\right) }^{-\rho }{E}_{i} + {o}_{p}\left( {A\left( {n/k}\right) }\right) \]\n\n\[ = \gamma {E}_{i} + A\left( {n/k}\...
Yes
Lemma 3.1 Suppose Assumption G holds. Then there exist constants \( {M}_{p},{M}_{y},{M}_{m} > 0 \) , such that\n\n\[ \begin{Vmatrix}{p}_{k}\end{Vmatrix} \leq {M}_{p},\;\begin{Vmatrix}{y}_{k + 1}\end{Vmatrix} \leq {M}_{y},\;\left| {{m}_{k}\left( \alpha \right) }\right| \leq {M}_{m}\alpha \]\n\nfor all \( k \) and \( \al...
Proof From (G1) we have that the right-hand side of (2.3) is uniformly bounded. Additionally, Assumptions (G2), (G3) and (G4) guarantee that the inverse of the matrix in (2.3) exists and is uniformly bounded for all \( k \) . Consequently, the solution of \( \left( {2.3}\right) \left( {\left( {p}_{k},{y}_{k + 1}\right)...
Yes
Lemma 3.2 Suppose Assumption (G1) holds. Then there exist constants \( {C}_{\theta },{C}_{\omega } > 0 \) such that for all \( k \) and \( \alpha \in (0,1\rbrack \)\n\n\[ \left| {\theta \left( {{x}_{k} + \alpha {p}_{k}}\right) - \left( {1 - \alpha }\right) \theta \left( {x}_{k}\right) }\right| \leq {C}_{\theta }{\alpha...
Proof The proof of (3.4a) can be found in Lemma 3.3 from [14].\n\nNext, we give the proof of (3.4b). From the second order Taylor expansions,\n\n\[ \left| {\omega \left( {{x}_{k} + \alpha {p}_{k}}\right) - \omega \left( {x}_{k}\right) - {m}_{k}\left( \alpha \right) }\right| \]\n\n\[ \overset{\left( {2.9}\right) }{ = }\...
No
Lemma 3.3 Suppose Assumption G holds. If \( \left\{ {x}_{{k}_{i}}\right\} \) is a subsequence of iterates for which \( \chi \left( {x}_{{k}_{i}}\right) \geq {\epsilon }_{1} \) with constants \( {\epsilon }_{1} > 0 \) independent of \( i \) . Then there exists \( {\epsilon }_{2} > 0 \) independent of \( i \), we have\n\...
Proof If \( \theta \left( {x}_{{k}_{i}}\right) = 0 \), but \( \chi \left( {x}_{{k}_{i}}\right) > 0 \), Algorithm I would not terminated. From (2.3) and (3.3),\n\n\[ {m}_{{k}_{i}}\left( \alpha \right) /\alpha = {\left( {g}_{{k}_{i}} - {A}_{k}{y}_{{k}_{i} + 1}\right) }^{\mathrm{T}}{H}_{{k}_{i}}{p}_{{k}_{i}} = - {\begin{V...
Yes
Lemma 3.4 Suppose Assumption G holds. Then the trial point \( {x}_{k}\left( {\alpha }_{k, l}\right) \) could not be rejected by \( {x}_{k} \) if \( {\alpha }_{k, l} \) is sufficiently small.
Proof There are two cases.\n\nCase \( 1\;\left( {\theta \left( {x}_{k}\right) > 0}\right) \) . Following directly from the second order Taylor expansion of \( \theta \left( x\right) \) , we have\n\n\[ \theta \left( {{x}_{k}\left( {\alpha }_{k, l}\right) }\right) - \theta \left( {x}_{k}\right) + \mu \left( {\alpha }_{k,...
Yes
Theorem 3.1 Suppose Assumption G holds. Then\n\n\[ \mathop{\lim }\limits_{{k \rightarrow \infty }}\theta \left( {x}_{k}\right) = 0 \]
Proof The proof is similar to Lemma 8 in [15].
No
Lemma 3.7 Suppose Assumption G holds. Let \( \left\{ {x}_{{k}_{i}}\right\} \) be a subsequence. There exists a certain constant \( \bar{\alpha } > 0 \) such that for all \( {k}_{i} \) and \( \alpha \leq \bar{\alpha } \)\n\n\[ \omega \left( {{x}_{{k}_{i}} + \alpha {p}_{{k}_{i}}}\right) - \omega \left( {x}_{{k}_{i}}\righ...
Proof Let \( {M}_{p} \) and \( {C}_{\omega } \) be the constants from Lemmas 3.1 and 3.2. It then follows for all \( \alpha \leq \bar{\alpha } \) with \( \bar{\alpha } \mathrel{\text{:=}} \frac{\left( {1 - {\eta }_{\omega }}\right) {\epsilon }_{2}}{{C}_{\omega }{M}_{p}^{2}} \) and from (3.5) that\n\n\[ \omega \left( {{...
Yes
Lemma 3.8 Suppose that Assumption G holds and that the filter is augmented only a finite number of times, i.e., \( \left| \mathcal{Z}\right| < \infty \) . Then\n\n\[ \mathop{\lim }\limits_{{k \rightarrow \infty }}\chi \left( {x}_{k}\right) = 0 \]
The proof of Lemma 3.8 is based on Lemma 8 in [4] by using the new filter pair.
No
Lemma 3.9 Suppose Assumption G holds. Let \( \left\{ {x}_{{k}_{i}}\right\} \) be a subsequence and \( {m}_{{k}_{i}}\left( \alpha \right) \leq \) \( - \alpha {\epsilon }_{2} \) for a constant \( {\epsilon }_{2} > 0 \) independent of \( {k}_{i} \) and for all \( \alpha \in (0,1\rbrack \) . Then there exist constants \( {...
The proof of Lemma 3.9 is similar to Lemma 9 in [4]. So, it is omitted.
No
Lemma 3.10 Suppose Assumption G holds. Let \( \left\{ {x}_{{k}_{i}}\right\} \) be a subsequence of \( \left\{ {x}_{k}\right\} \) with \( \chi \left( {x}_{{k}_{i}}\right) \geq \epsilon \) for a constant \( \epsilon > 0 \) independent of \( {k}_{i} \) . Then there exists \( K \in \mathbb{N} \) such that for all \( {k}_{i...
Lemma 3.10 can be proved by using the idea of Lemma 10 in [4].
No
Theorem 3.2 Suppose Assumption G holds. Then\n\n\\[ \n\\mathop{\\lim }\\limits_{{k \\rightarrow \\infty }}\\theta \\left( {x}_{k}\\right) = 0 \n\\]\n\n\\( \\left( {{3.11}\\mathrm{a}}\\right) \\)\n\nand\n\n\\[ \n\\mathop{\\liminf }\\limits_{{k \\rightarrow \\infty }}\\chi \\left( {x}_{k}\\right) = 0 \n\\]\n\n(3.11b)\n\n...
Proof (3.11a) follows from Theorem 3.1. In the case of the filter is augmented only a finite number of times, (3.11b) has been proved by Lemma 3.8. Otherwise, there exists a subsequence \\( \\left\\{ {x}_{{k}_{i}}\\right\\} \\) such that \\( {k}_{i} \\in \\mathcal{Z} \\) for all \\( i \\) . Now suppose that \\( \\lim \...
Yes
Theorem 2.1 For all \( f \in {L}_{{p}_{2}\left( \cdot \right) ,{q}_{2}}, g \in {L}_{p\left( \cdot \right), q},0 < {p}^{ + },{p}_{2}^{ + } < \infty ,0 < q,{q}_{2} \leq \infty \), with \( \frac{1}{{p}_{1}} = \frac{1}{p} + \frac{1}{{p}_{2}},\frac{1}{{q}_{1}} = \frac{1}{q} + \frac{1}{{q}_{2}} \), we have
\[ \parallel {fg}{\parallel }_{{L}_{{p}_{1}\left( \cdot \right) ,{q}_{1}}} \leq c\parallel f{\parallel }_{{L}_{{p}_{2}\left( \cdot \right) ,{q}_{2}}}\parallel g{\parallel }_{{L}_{p\left( \cdot \right), q}}. \]
Yes
Theorem 3.1 Let \( p\\left( \\cdot \\right) ,{p}_{2}\\left( \\cdot \\right) \\in P\\left( \\Omega \\right) \) satisfy the condition (2.4), \( 0 < q,{q}_{2} < \\infty, v \\in \) \( {V}_{p\\left( \\cdot \\right), q} \) with \( \\frac{1}{{p}_{1}\\left( \\cdot \\right) } = \\frac{1}{p\\left( \\cdot \\right) } + \\frac{1}{{...
Proof Using the pointwise estimation\n\n\[ s\\left( {{T}_{v}\\left( f\\right) }\\right) \\leq M\\left( v\\right) s\\left( f\\right) \]\n\n(3.1)\n\nby Hölder's inequality in Theorem 2.1, we have\n\n\[ {\\begin{Vmatrix}s\\left( {T}_{v}\\left( f\\right) \\right) \\end{Vmatrix}}_{{p}_{1}\\left( \\cdot \\right) ,{q}_{1}} \\...
Yes
Lemma 4. \( {\mathbf{1}}^{\left\lbrack {17}\right\rbrack } \) Let \( p\left( \cdot \right) \in P\left( \Omega \right) ,0 < q \leq \infty ,0 < \theta < 1 \) and \( \frac{1}{p\left( \cdot \right) } = \frac{1 - \theta }{\widehat{p}\left( \cdot \right) } \) . Then
\[ {\left( {H}_{\widehat{p}\left( \cdot \right) }^{s},{H}_{\infty }^{s}\right) }_{\theta, q} = {H}_{p\left( \cdot \right), q}^{s}. \]
Yes
Theorem 4.1 Let \( {p}_{1}\left( \cdot \right) ,{p}_{2}\left( \cdot \right) \in P\left( \Omega \right) ,0 < {p}_{1}\left( \cdot \right) < {p}_{2}\left( \cdot \right) < \infty \) and \( 0 < q < \infty \) . Suppose that \( f = {\left\{ {f}_{n}\right\} }_{n \geq 0} \in {H}_{{p}_{1}\left( \cdot \right), q}^{s} \), one of i...
Proof From the definition of \( {v}_{j}^{-1} \), it is easy to see that the process \( {v}^{-1} = {\left\{ {v}_{j}^{-1}\right\} }_{j \geq 1} \) is adapted to \( {\left\{ {\mathcal{F}}_{j}\right\} }_{j \geq 1} \) and \( {v}_{j}^{-1} \leq 1 \) for every \( j \geq 1 \) . Then \( g = {\left\{ {g}_{n}\right\} }_{n \geq 0} \...
Yes
Theorem 4.2 Let \( {p}_{1}\left( \cdot \right) ,{p}_{2}\left( \cdot \right) \in P\left( \Omega \right) ,0 < {p}_{1}\left( \cdot \right) < {p}_{2}\left( \cdot \right) < \infty \) and \( 0 < {q}_{1} < {q}_{2} < \infty \) . Suppose that \( f = {\left\{ {f}_{n}\right\} }_{n \geq 0} \in {H}_{{p}_{1}\left( \cdot \right) ,{q}...
\[ {g}_{n} = \mathop{\sum }\limits_{{k = 0}}^{n}{v}_{k - 1}^{-1}d{f}_{k},\;n \geq 0, \] where \( {v}_{j}^{-1} \mathrel{\text{:=}} \min \left\{ {\mathrm{E}\left( {{s}_{j + 1}{\left( {f}^{0}\right) }^{-\beta } \mid {\mathcal{F}}_{j}}\right) ,1}\right\} \) for any \( j \geq - 1,{f}^{0} \) is given by (4.2) and \( \beta = ...
Yes
For the MARE (1.1), if \( K \) in (1.2) is a nonsingular M-matrix or an irreducible singular M-matrix and the parameters \( \alpha ,\beta \) satisfy\n\n\[ \alpha \geq \max \left\{ {a}_{ii}\right\} ,\;\beta \geq \max \left\{ {d}_{ii}\right\} \]\n\n(3.5)\n\nthen the sequence \( \left\{ {X}_{k}\right\} \) generated by (3....
Proof We prove (3.6) by induction. When \( k = 0 \), it is clear that \( 0 = {X}_{0} \leq S \) . Thus, (3.6) holds true for \( k = 0 \) .\n\nSuppose that (3.6) holds true for \( k = l \) . Then we have\n\n\[ \left( {{\beta I} + A - {X}_{l}C}\right) \left( {{X}_{l + 1} - S}\right) \left( {{\alpha I} + D - C{X}_{l}}\righ...
Yes
Theorem 3.1 For the MARE (1.1), if \( K \) in (1.2) is a nonsingular M-matrix or an irreducible singular M-matrix and the parameters \( \alpha ,\beta \) satisfy (3.5), then the sequence \( \left\{ {X}_{k}\right\} \) generated by the new iteration method (3.4) is well defined, monotonically increasing, and converges to ...
Proof We have shown in Lemma 3.1 and Lemma 3.2 that \( \left\{ {X}_{k}\right\} \) is nonnegative, monotonically increasing and is bounded from above. Thus there is a nonnegative matrix \( {S}^{ * } \) such that \( \mathop{\lim }\limits_{{k \rightarrow \infty }}{X}_{k} = {S}^{ * } \) . From Lemma 3.1, we have \( {S}^{ *...
Yes
Example 4.1 Consider the MARE (1.1) with\n\n\[ A = - {10}{E}_{n \times n} + {180.002}{I}_{n},\;B = {0.001}{E}_{n \times m}, \]\n\n\[ C = {B}^{\mathrm{T}},\;D = {0.018}{I}_{m}, \]\n\nwhere \( {E}_{m \times n} \) is the \( m \times n \) matrix with all ones and \( {I}_{m} \) is the identity matrix of size \( m \) with \(...
Tab. 4.1 Numerical results of Example 4.1\n\n<table><thead><tr><th>Method</th><th>IT</th><th>CPU</th><th>RES</th></tr></thead><tr><td>Newton</td><td>3</td><td>0.0022</td><td>\( {7.4339}\mathrm{e} \) - \( {08} \)</td></tr><tr><td>FP3</td><td>8</td><td>0.0043</td><td>\( {4.8065}\mathrm{e} \) - \( {07} \)</td></tr><tr><td...
No
Example 4.2 Consider the MARE (1.1) with\n\n\[ A = \left( \begin{matrix} 3 & - 1 & & \\ & 3 & \ddots & \\ & & \ddots & - 1 \\ - 1 & & & 3 \end{matrix}\right) \in {\mathbb{R}}^{n \times n},\;B = 2{I}_{n},\;C = {20}{I}_{n},\;D = {10A}. \]
This example is from [12], where the corresponding \( K \) is an irreducible singular M-matrix. For different sizes of \( n \), the numerical results are summarized in Tab. 4.2.\n\nTab. 4.2 Numerical results of Example 4.2\n\n<table><thead><tr><th>\( n \)</th><th>Method</th><th>IT</th><th>CPU</th><th>RES</th></tr></the...
No
Theorem 1.1 Suppose \( u \) satisfies \( {u}_{xx} + {x}^{2\sigma }{u}_{yy} = f \) in \( \Omega \) and \( u\left( {0, y}\right) = \varphi \left( y\right), u\left( {\kappa, y}\right) = \phi \left( y\right) \), then the following holds.
1) Let \( 0 < \alpha < 1 \) when \( k = 0,1 \) . If \( f\left( {x, y}\right) \in {L}^{\infty }\left( \Omega \right) ,\varphi \left( y}\right) \in {C}_{ * }^{k,\alpha }\left( \mathbb{R}\right) \) and \( \psi \left( y}\right) \in {C}^{k,\alpha }\left( \mathbb{R}\right) \) , then \( u \in {C}_{ * }^{k,\alpha }\left( \bar{...
Yes
Lemma 3.1 Suppose \( \Omega \) is a bounded domain in \( {\mathbb{R}}_{ + }^{2} \), and \( \mathcal{L}u \leq 0 \) in \( \Omega, u \geq 0 \) on \( \partial \Omega \) , then \( u \geq 0 \) in \( \Omega \) .
It is easily seen by applying the usual methods for the elliptic equations, so we omit the proof.
No
Lemma 3.2 Let \( Q = \\left( {0,1}\\right) \\times \\left( {-2,2}\\right) \) . Suppose\n\n\[ \n{u}_{xx} + {x}^{2\\sigma }{u}_{yy} = 0,\\;\\left( {x, y}\\right) \\in Q = \\left( {0,1}\\right) \\times \\left( {-2,2}\\right) \n\]\n\nwith \( u\\left( {0, y}\\right) = 0 \), and \( \\left| u\\right| \\leq 1,\\left( {x, y}\\r...
Proof By Lemma 3.1 we know\n\n\[ \n\\left| u\\right| \\leq 1,\\;\\left( {x, y}\\right) \\in Q. \n\]\n\nWe consider the function\n\n\[ \nv\\left( {x, y}\\right) = - {x}^{2} + {y}^{2} + {2x} \n\]\n\nthen\n\n\[ \n{v}_{xx} + {x}^{2\\sigma }{v}_{yy} = - 2 + 2{x}^{2\\sigma } \\leq 0,\\;\\left( {x, y}\\right) \\in {Q}_{1}^{ +...
Yes
Lemma 3.3 Let\n\n\\[ \n{u}_{xx} + {x}^{2\\sigma }{u}_{yy} = f,\\;\\left( {x, y}\\right) \\in {Q}_{1}^{ + },\n\\]\n\n(3.1)\n\nand \\( u\\left( {0, y}\\right) = \\varphi \\left( y\\right), y \\in \\left( {-1,1}\\right) \\) . There exist constants \\( 0 < r < \\frac{1}{2} \\) and \\( \\delta \\) small such that if\n\n\\[\...
Proof Let \\( v\\left( {x, y}\\right) \\) satisfy\n\n\\[\n\\left\\{ \\begin{array}{ll} {v}_{xx} + {x}^{2\\sigma }{v}_{yy} = 0, & \\left( {x, y}\\right) \\in {Q}_{1}^{ + }, \\\\ v\\left( {0, y}\\right) = 0, & y \\in \\left( {-1,1}\\right) , \\\\ v = u, & \\left( {x, y}\\right) \\in \\partial {Q}_{1}^{ + } \\smallsetminu...
Yes
Lemma 3.4 Suppose \( u \) is a weak solution of\n\n\[ \n{u}_{xx} + {x}^{2\sigma }{u}_{yy} = 0,\;\left( {x, y}\right) \in {Q}_{1}^{ + }, \]\n\nwith the boundary condition\n\n\[ \nu\left( {0, y}\right) = 0,\;y \in \left( {-1,1}\right) \]\n\nand \( \left| {u\left( {x, y}\right) }\right| \leq 1,\left( {x, y}\right) \in {Q}...
This lemma can be proved by the odd extension and Lemma 2 in [10].
No
Lemma 3.5 Suppose\n\n\\[ \n{u}_{xx} + {x}^{2\\sigma }{u}_{yy} = f,\\;\\left( {x, y}\\right) \\in {Q}_{1}^{ + },\n\\]\n\nand \\( u\\left( {0, y}\\right) = \\varphi \\left( y\\right) ,\\left| u\\right| \\leq 1,\\left( {x, y}\\right) \\in \\partial {Q}_{1}^{ + } \\) . Then, for any \\( 0 < \\alpha < 1 \\), there exist con...
Proof Let \\( v\\left( {x, y}\\right) \\) satisfy\n\n\\[ \n\\left\\{ \\begin{array}{ll} {v}_{xx} + {x}^{2\\sigma }{v}_{yy} = 0, & \\left( {x, y}\\right) \\in {Q}_{1}^{ + }, \\\\ v\\left( {0, y}\\right) = 0, & y \\in \\left( {-1,1}\\right) , \\\\ v = u, & \\left( {x, y}\\right) \\in \\partial {Q}_{1}^{ + } \\smallsetmin...
Yes
Lemma 4.1 Let \( A \geq 0 \) and \( 0 < \delta \leq 1 \) . Suppose\n\n\[ \n\mathcal{L}u = f,\;\left( {x, y}\right) \in {Q}_{\frac{1}{2}}^{ + }, \n\]\n\nwith\n\n\[ \nu\left( {0, y}\right) = \varphi \left( y\right) ,\;y \in \left( {-{\left( \frac{1}{2}\right) }^{1 + \sigma },{\left( \frac{1}{2}\right) }^{1 + \sigma }}\ri...
Proof Let\n\n\[ \nv\left( {x, y}\right) = \frac{1 + A}{2}\delta \left( {1 - 4{x}^{2}}\right) . \n\]\n\nThen we have\n\n\[ \nv\left( {x, y}\right) \geq u\left( {x, y}\right) ,\;\left( {x, y}\right) \in \partial {Q}_{\frac{1}{2}}^{ + } \n\]\n\nand\n\n\[ \n\mathcal{L}v = - 4\left( {1 + A}\right) \delta - 4\left( {1 + A}\r...
Yes
Lemma 4.2 Suppose\n\n\\[ \n\\mathcal{L}u = f,\\;\\left( {x, y}\\right) \\in {Q}_{1}^{ + },\n\\]\n\nwith \\( u\\left( {0, y}\\right) = \\varphi \\left( y\\right) \\), and \\( \\left| u\\right| \\leq 1,\\left( {x, y}\\right) \\in \\partial {Q}_{1}^{ + } \\) . Then, for any \\( 0 < \\alpha < 1 \\), there exist constants \...
Proof Let \\( v\\left( {x, y}\\right) \\) satisfy\n\n\\[ \n\\left\\{ \\begin{array}{ll} {v}_{xx} + {x}^{2\\sigma }{v}_{yy} = 0, & \\left( {x, y}\\right) \\in {Q}_{1}^{ + }, \\\\ v\\left( {0, y}\\right) = 0, & y \\in \\left( {-1,1}\\right) , \\\\ v = u, & \\left( {x, y}\\right) \\in \\partial {Q}_{1}^{ + } \\smallsetmin...
Yes
Theorem 4.1 Let \( l > \sigma - 1,0 < \alpha < 1,\varphi \left( y\right) \) be \( {C}_{ * }^{1,\alpha } \) at \( y = 0 \) . Suppose\n\n\[ \left\{ \begin{array}{ll} \mathcal{L}u = f, & \left( {x, y}\right) \in {Q}_{{r}_{0}}^{ + }, \\ u\left( {0, y}\right) = \varphi \left( y\right) , & y \in \left( {-{r}_{0}^{1 + \sigma ...
Proof We will modify the proof of Theorem 3.1 to obtain this theorem. Since \( \varphi \left( y\right) \) is \( {C}_{ * }^{1,\alpha } \) at \( y = 0 \), we have\n\n\[ \left| {\varphi \left( y\right) - p\left( y\right) }\right| \leq {\left\lbrack \varphi \right\rbrack }_{{C}_{ * }^{1,\alpha }\left( 0\right) }{y}^{\frac{...
Yes
Theorem 2.1 Let \( \mathcal{A} = W + \mathrm{i}T \in {\mathbb{C}}^{n \times n} \) be a non-Hermitian but symmetric matrix \( \left( {\mathcal{A} \neq {\mathcal{A}}^{ * },\mathcal{A} = {\mathcal{A}}^{\mathrm{T}}}\right) \) with \( W, T \in {\mathbb{R}}^{n \times n} \) are both symmetric, and \( W \) being positive defin...
Proof (i) By (2.4) and direct calculations, we have\n\n\[ \n\rho \left( {T}_{\alpha ,\beta }\right) = \rho \left( {\mathrm{i}{\widehat{M}}_{\alpha ,\beta }\left( {{\beta I} - \alpha {W}^{-1}T}\right) }\right) \n\]\n\n\[ \n\leq {\begin{Vmatrix}\mathrm{i}{\widehat{M}}_{\alpha ,\beta }\left( \beta I - \alpha {W}^{-1}T\rig...
Yes
Corollary 2.1 Assume that the conditions of Theorem 2.1 are satisfied, then the optimal relation between two parameters \( \alpha ,\beta \) that minimizes the upper bound \( {\delta }_{\alpha ,\beta } \) of the spectral radius \( \rho \left( {T}_{\alpha ,\beta }\right) \) is given by\n\n\[ \alpha = \beta \frac{1 - {\la...
Proof By introducing \( \tau = \alpha /\beta \), and\n\n\[ g\left( \tau \right) = \frac{1 - \tau {\lambda }_{\min }}{\tau + {\lambda }_{\min }},\;h\left( \tau \right) = \frac{\tau {\lambda }_{\max } - 1}{\tau + {\lambda }_{\max }}, \]\n\nwe have\n\n\[ {g}^{\prime }\left( \tau \right) = - \frac{1 + {\lambda }_{\min }^{2...
Yes
Theorem 2.2 Let \( \mathcal{A} = W + \mathrm{i}T \in {\mathbb{C}}^{n \times n} \) be a non-Hermitian but symmetric matrix \( \left( {\mathcal{A} \neq {\mathcal{A}}^{ * },\mathcal{A} = {\mathcal{A}}^{\mathrm{T}}}\right) \) with \( W, T \in {\mathbb{R}}^{n \times n} \) be both symmetric, and \( W \) being positive defini...
Proof Let \( \lambda \) be an eigenvalue of the matrix \( {T}_{\alpha ,\beta } \) and \( x \) the corresponding eigenvector, i.e., \( {M}_{\alpha ,\beta }^{-1}{N}_{\alpha ,\beta }x = {\lambda x} \), or equivalently, \( \lambda \left( {{\alpha W} + {\beta T}}\right) x = \mathrm{i}\left( {{\beta W} - {\alpha T}}\right) x...
Yes
Theorem 2.3 Let \( \mathcal{A} = W + \mathrm{i}T \in {\mathbb{C}}^{n \times n} \) be a non-Hermitian but symmetric matrix \( \left( {\mathcal{A} \neq {\mathcal{A}}^{ * },\mathcal{A} = {\mathcal{A}}^{\mathrm{T}}}\right) \) with \( W, T \in {\mathbb{R}}^{n \times n} \) be both symmetric, and \( W \) being positive defini...
Proof Let \( \lambda \) be an eigenvalue of the matrix \( {M}_{\alpha ,\beta }^{-1}\mathcal{A} \) and \( x \) the corresponding eigenvector with \( \parallel x{\parallel }_{2} = 1 \) . It is known that\n\n\[ \left( {\alpha - \beta \mathrm{i}}\right) \left( {W + \mathrm{i}T}\right) x = \lambda \left( {{\alpha W} + {\bet...
Yes
Lemma 2.1 Systems (1.2) and (1.3) have a unique global positive solution on \( t > - \tau \) for any initial data given above, respectively.
Remark 2.1 The proof is very standard and is omitted here. Readers may refer to [17].
No
Theorem 3.3 If Assumption 3.2 holds, then the equilibrium state of (3.5) is stable in probability, i.e., the equilibrium state of (1.3) is stable in probability.
Proof Define two same functionals \( {V}_{1},{V}_{2} \) as before, using Itô’s formula and computing \( {LV} \) along (3.6), we have\n\n\[ L\left( {{V}_{1} + {V}_{2}}\right) = 2\left( {{y}_{1} - \frac{{a}_{12}{x}_{1}^{ * }}{1 - {\tau }_{1}^{\prime }}{\int }_{t - {\tau }_{1}\left( t\right) }^{t}{y}_{1}\left( s\right) \m...
Yes
Theorem 2.1 Let Assumptions 1-3 hold. Then, for small enough \( \varepsilon > 0 \) there exists a \( {\lambda }^{ * }\left( \varepsilon \right) \) with \( {\lambda }^{ * }\left( \varepsilon \right) = {\lambda }_{1}\varepsilon + \mathcal{O}\left( {\varepsilon }^{3/2}\right) \) such that when \( \lambda = {\lambda }^{ * ...
## 3. Proof of Theorem 2.1\n\nTo prove Theorem 2.1 we need the following lemma whose proof can be found in [24].\n\nLemma 3.1 \( {}^{\left
No
Lemma 3.2 Assume \( \theta \in \left( {0, H}\right), H \in \left( {\frac{1}{2},1}\right) \), and let \( {\xi }_{t} \) be defined in (3.1). Then, as \( t \rightarrow T \) :\n\n1) if \( 0 < \theta < \frac{1}{2} \), then\n\n\[ \n{\left( T - t\right) }^{1 - {2\theta }}{\int }_{0}^{t}{\left( T - u\right) }^{{2\theta } - 2}{...
Proof of Theorem 3.1 By the formula (2.6), we obtain that, for any \( t \in \lbrack 0, T) \) ,\n\n\[ \n\frac{1}{2}{\left( T - t\right) }^{{2\theta } - 1}{\xi }_{t}^{2} = \frac{1 - {2\theta }}{2}{\int }_{0}^{t}{\left( T - u\right) }^{{2\theta } - 2}{\xi }_{u}^{2}\mathrm{\;d}u + {\int }_{0}^{t}{\left( T - u\right) }^{{2\...
Yes
1) If \( \theta \in \left( {0,1 - H}\right) \) then, as \( t \rightarrow T \) ,\n\n\[ \left( {F,{\left( T - t\right) }^{1 - H - \theta }{\int }_{0}^{t}{\left( T - u\right) }^{\theta - 1}\mathrm{\;d}{Y}_{u}^{\left( 1\right) }}\right) \overset{\text{ law }}{ \rightarrow }\left( {F,\sqrt{{\sigma }_{1}}N}\right) . \]
Proof of Theorem 3.2 1) Assume that \( \theta \in \left( {0,1 - H}\right) \) . By Lemma 3.4, we have\n\n\[ {\left( T - t\right) }^{\theta - H}\left( {\theta - {\widehat{\theta }}_{t}}\right) = \frac{{\left( T - t\right) }^{1 - H - \theta }{\eta }_{t}}{{\left( T - t\right) }^{1 - {2\theta }}{\int }_{0}^{t}{\left( T - u\...
Yes
Theorem 2.1 The Pitman's measure of closeness of preliminary test estimator related to unrestricted estimator are as follows:\n\n(i) If \( {F}_{\alpha } \leq \frac{m\left( {p + \delta - \gamma }\right) }{p\left( {m + \gamma }\right) } \), then\n\n\[ \operatorname{PC}\left( {{\widehat{\sigma }}^{2PT},{\widehat{\sigma }}...
Proof By direct computation, we know that\n\n\[ L\left( {{\widehat{\sigma }}^{2PT},{\sigma }^{2}}\right) = {\left( \frac{{\widehat{\sigma }}^{2PT}}{{\sigma }^{2}} - 1\right) }^{2} = I\left( {F \geq {F}_{\alpha }}\right) {\left( \frac{{u}_{2}}{m + \gamma } - 1\right) }^{2} + I\left( {F < {F}_{\alpha }}\right) {\left( \f...
Yes
Example 1 Symmetry reduction and solution of the system (2.1) which admits the GCS (2.2).
We get solutions in exponential form by integrating the system (2.2)\n\n\[ u = {\phi }_{1}\left( t\right) + {\phi }_{2}\left( t\right) {\mathrm{e}}^{{b}_{1}x},\;v = {\psi }_{1}\left( t\right) + {\psi }_{2}\left( t\right) {\mathrm{e}}^{{b}_{1}x}. \]\n\n(3.1)\n\nThe system (2.1) can be reduced to the following system of ...
Yes
The system (2.9) can be rewritten as\n\n\[ \n{u}_{t} = {\left\lbrack \left( d + av\right) u\right\rbrack }_{xx} + u\left( {{a}_{11} - {4a}{c}^{2}v}\right) ,\;{v}_{t} = {\left\lbrack \left( d + bu\right) v\right\rbrack }_{xx} + v\left( {{a}_{21} - {4b}{c}^{2}u}\right) .\n\] \n\n(3.3)
Since \( a > 0, b > 0, d > 0,{a}_{11} > 0,{a}_{21}{c}^{2} > 0 \), the system (3.3) is competition model. The form of GCSs is\n\n\[ \n{\eta }_{1} = {u}_{xxx} - {c}^{2}{u}_{x},\;{\eta }_{2} = {v}_{xxx} - {c}^{2}{v}_{x}.\n\] \n\n(3.4)\n\nWe obtain the exponential form solutions by integrating the system (3.4)\n\n\[ \nu = ...
Yes
Theorem 3.1(Weak duality theorem) Let \( x \in M \) be feasible for \( \left( \mathcal{P}\right) \) and \( \left( {\lambda ,\rho }\right) (\lambda \in \) \( {T}^{ * },\rho \geq 0) \) be a feasible point for \( \left( \mathcal{D}\right) \), then\n\n\[ \operatorname{val}\left( \mathcal{D}\right) \leq \operatorname{val}\l...
Proof As \( \mathcal{L}\left( {x,\lambda ,\rho }\right) \leq \psi \left( {x,0}\right) \), it follows that\n\n\[ \mathop{\inf }\limits_{{x \in M}}\mathcal{L}\left( {x,\lambda ,\rho }\right) \leq \psi \left( {x,0}\right) \]\n\ntherefore\n\n\[ \mathop{\inf }\limits_{{x \in M}}\mathcal{L}\left( {x,\lambda ,\rho }\right) \l...
Yes
Corollary 3.2 Suppose the condition (Q) holds, and \( \mu \) is lower semi-continuous at 0 . then \( \operatorname{val}\left( \mathcal{P}\right) = \operatorname{val}\left( \mathcal{D}\right) \) .
Proof According to the condition (Q), there exist real constants \( c \) and \( b \), such that \( \mu \left( z\right) \geq c - {b\sigma }\left( z\right) ,\forall z \in T. \)\n\nLet \( \lambda \in {T}^{ * } \), then\n\n\[ \mu \left( z\right) \geq c + \langle \lambda, z\rangle - \parallel \lambda {\parallel }_{1}\parall...
Yes
Example 3.1 Consider the following problem:\n\n\\[ \n\\text{(}\\mathcal{P}\\text{)}\\min f\\left( x\\right) = {x}_{1}^{2} - {x}_{2}^{2} + 2{x}_{2} \n\\]\n\n\\[ \n\\text{s.t}\\;g\\left( {x,\\omega }\\right) = {x}_{1} + \\left( {1 - \\omega }\\right) {x}_{2} \\leq 0\\text{,}\n\\]\n\n(3.9)\n\n\\[ \n\\omega \\in \\left\\lb...
As \\( \\mathcal{L}\\left( {x,\\lambda ,\\rho }\\right) \\leq \\psi \\left( {x,0}\\right) \\), it follows that\n\n\\[ \n\\mathop{\\inf }\\limits_{{x \\in M}}\\mathcal{L}\\left( {x,\\lambda ,\\rho }\\right) \\leq \\psi \\left( {x,0}\\right) \n\\]\n\ntherefore\n\n\\[ \n\\mathop{\\inf }\\limits_{{x \\in M}}\\mathcal{L}\\l...
Yes
We consider the case of the LAD loss function and the error term with a standard normal distribution. The estimated results are shown in Tab. 1.
From Tab. 1 we find that FR+LLA method performs best in the case of ultra-high dimensional data. The RSIS+Lasso method performs better for the estimation error and the prediction error. However, considering the ability to correctly exclude unimportant variables and to misidentify important variables, the FR+LLA method ...
Yes
Lemma 2.2 Let \( {x}^{ * } \in {X}^{ * } \) . Then, the sequences \( \left\{ {x}_{k}\right\} \) and \( \left\{ {z}_{k}\right\} \) generated by Algorithm SGM satisfy\n\n\[ \n{\begin{Vmatrix}{x}_{k + 1} - {x}^{ * }\end{Vmatrix}}^{2} \leq {\begin{Vmatrix}{x}_{k} - {x}^{ * }\end{Vmatrix}}^{2} - {\sigma }^{2}{\begin{Vmatrix...
Proof See the proof of Lemma 3.2 in [8].
No
Lemma 3.1 Suppose that Assumption (A2) holds. Then, there exists a constant \( \gamma > 0 \) such that the stepsize \( {\alpha }_{k} \) defined in Step 3 of Algorithm SGM satisfies\n\n\[ \n{\alpha }_{k} \geq \min \left\{ {1,\gamma \frac{{\begin{Vmatrix}{F}_{k}\end{Vmatrix}}^{2}}{{\begin{Vmatrix}{d}_{k}\end{Vmatrix}}^{2...
Proof See the proof of Lemma 3.4 in [8].
No
Lemma 3.2 Let \( \\left\\{ {x}_{k}\\right\\} \) be an infinite sequence generated by Algorithm SGM. If Assumption (A2) holds, then we have\n\n\[ \n\\mathop{\\lim }\\limits_{{k \\rightarrow + \\infty }}\\inf \\begin{Vmatrix}{F}_{k}\\end{Vmatrix} = 0 \n\]\n\nFurthermore, the sequence \( \\left\\{ {x}_{k}\\right\\} \) con...
Proof See the proof of Theorem 3.5 in [8].
No
Theorem 3.1 Suppose that Assumptions (A1) and (A2) hold. Then, there exists a constant \( \tau > 0 \) such that\n\n\[ \operatorname{dist}\left( {{x}_{k},{X}^{ * }}\right) \leq \frac{\tau }{\sqrt{k}} \]\n\n(3.7)
Proof Let \( {\bar{x}}_{k} \in {X}^{ * } \) be the closest point to \( {x}_{k} \), i.e.,\n\n\[ \begin{Vmatrix}{{x}_{k} - {\bar{x}}_{k}}\end{Vmatrix} = \operatorname{dist}\left( {{x}_{k},{X}^{ * }}\right) \]\n\n(3.8)\n\nThen, it follows from (2.6) and (3.8) that\n\n\[ {\operatorname{dist}}^{2}\left( {{x}_{k + 1},{X}^{ *...
Yes
Theorem 3.2 Suppose that Assumption (A2) holds. If the mapping \( F \) is strongly monotone with modulus \( \mu > 0 \), i.e., \n\n\[ \n\langle F\left( x\right) - F\left( y\right), x - y\rangle \geq \mu \parallel x - y{\parallel }^{2},\forall x, y \in X. \n\] \n\nthen \n\n\[ \n\begin{Vmatrix}{{x}_{k} - \bar{x}}\end{Vmat...
Proof By the Cauchy-Schwarz inequality and the strong monotonicity of \( F \), we obtain \n\n\[ \n\begin{Vmatrix}{F}_{k}\end{Vmatrix} = \begin{Vmatrix}{{F}_{k} - F\left( \bar{x}\right) }\end{Vmatrix} \geq \mu \begin{Vmatrix}{{x}_{k} - \bar{x}}\end{Vmatrix}, \n\] \n\n(3.16) \n\nwhich, together (3.10) and (3.11), implies...
Yes
Lemma 4.1 Let \( \left\{ {x}_{k}\right\} \) and \( \left\{ {z}_{k}\right\} \) be the two sequences generated by UAF. If \( {x}^{ * } \in {X}^{ * } \) , then we have the following conclusions:\n\n1) If the stepsize \( {\alpha }_{k} \) is performed by using the line search schemes (4.4),(4.5) and (4.9), respectively, the...
Proof From the monotonicity of \( F \) and \( {x}^{ * } \in {X}^{ * } \), it follows that\n\n\[ \nF{\left( {z}_{k}\right) }^{\mathrm{T}}\left( {{z}_{k} - {x}^{ * }}\right) = {\left( F\left( {z}_{k}\right) - F\left( {x}^{ * }\right) \right) }^{\mathrm{T}}\left( {{z}_{k} - {x}^{ * }}\right) \geq 0,\forall k.\n\]\n\n(4.13...
No
Lemma 4.2 Suppose that Assumption (A2) holds. Then, there exists a constant \( {c}_{10} > \) 0 such that the stepsize \( {\alpha }_{k} \) defined in Step 3 of UAF satisfies\n\n\[{\alpha }_{k} \geq \min \left\{ {\tau ,{c}_{10}\frac{{\begin{Vmatrix}{F}_{k}\end{Vmatrix}}^{2}}{{\begin{Vmatrix}{d}_{k}\end{Vmatrix}}^{2}}}\ri...
Proof The proof is similar to that of Lemma 2 in [12], and we therefore omit it.
No
Theorem 4.1 Let \( \\left\\{ {x}_{k}\\right\\} \) be an infinite sequence generated by UAF. If the stepsize \( {\\alpha }_{k} \) is performed by using the line search schemes (4.4), (4.5) and (4.9), respectively, then we have the following conclusions:\n\n(I) Suppose that Assumption (A2) holds. Then \( \\mathop{\\lim }...
Proof Using Lemma 4.11) and Lemma 4.2, we can prove this theorem in a similar way as Lemma 3.2, Theorem 3.1 and Theorem 3.2.
No
Theorem 4.2 Let \( \left\{ {x}_{k}\right\} \) be an infinite sequence generated by UAF. If the stepsize \( {\alpha }_{k} \) is performed by using the line search schemes (4.3) and (4.7), respectively, then we have the following conclusions:\n\n(I) Suppose that Assumption (A2) holds. Then \( \mathop{\lim }\limits_{{k \r...
Proof Using Lemma 4.12) and Lemma 4.2, we can prove this theorem in a similar way as Theorem 3.4, Theorem 3.6 and Theorem 3.8 in [7].
No
The elements of function \( F \) are given by\n\n\( {F}_{i}\left( x\right) = \ln \left( {{x}_{i} + 1}\right) - \frac{{x}_{i}}{n}, i = 1,2,\ldots, n \), and \( X = \left\{ {x \in {\mathbb{R}}^{n} \mid {x}_{i} \geq 0, i = 1,2,\ldots, n}\right\} \) .
Throughout the experiments, the direction \( {d}_{k} \) is defined as \( {}^{\left\lbrack 1\right\rbrack } \) :\n\n\[ \n{d}_{k} = \left\{ \begin{array}{ll} - {F}_{k}, & k = 0 \\ - {\theta }_{k}{F}_{k}, & k \geq 1 \end{array}\right. \]\n\n(5.1)\n\nwhere\n\n\[ \n{\theta }_{k} = \left\{ \begin{array}{ll} 1, & k = 0 \\ \fr...
Yes
Lemma 2.2 There exist \( m > 0 \) and \( C > 0 \) such that the solution of (1.5) fulfills\n\n\[{\int }_{\Omega }u\left( {\cdot, t}\right) \leq m\text{ for all }t \in \left( {0,{T}_{\max }}\right) ,\]\n\n(2.2)\n\nand\n\n\[{\int }_{t}^{t + \tau }{\int }_{\Omega }{u}^{\alpha } \leq C\text{ for all }t \in \left( {0,{T}_{\...
Proof The proof is analogous to Lemma 2.2 in [14]. To avoid repetition, we omit it.
No
Lemma 2.3 Suppose that \( \left( {u, v, w}\right) \) is a classical solution to (1.5). Then for each \( p \in \lbrack 1,\infty ) \), we have\n\n\[ \n{\int }_{\Omega }{v}^{p}\left( {\cdot, t}\right) \leq {\int }_{\Omega }{v}_{0}^{p}\text{ for all }t \in \left( {0,{T}_{\max }}\right) .\n\]\n\n(2.4)\n\nIn particular, when...
Proof Multiplying the second equation in (3.5) by \( {v}^{p - 1} \), integrating by parts, we obtain\n\n\[ \n\frac{1}{p}\frac{\mathrm{d}}{\mathrm{d}t}{\int }_{\Omega }{v}^{p} = {\int }_{\Omega }{v}^{p - 1}{\Delta v} - {\int }_{\Omega }{v}^{p}w\n\]\n\n\[ \n= - \left( {p - 1}\right) {\int }_{\Omega }{v}^{p - 2}{\left| \n...
Yes
Lemma 2.4 There exists \( C > 0 \) such that the solution of (1.5) satisfies\n\n\[ \mathop{\sup }\limits_{{t \in \left( {0,{T}_{\max }}\right) }}\parallel w\left( {\cdot, t}\right) {\parallel }_{{L}^{\alpha }\left( \Omega \right) } \leq C. \]
Proof Multiplying the third equation of (1.5) by \( {w}^{\alpha - 1} \) and using Young’s inequality, we obtain that\n\n\[ \frac{1}{\alpha }\frac{\mathrm{d}}{\mathrm{d}t}{\int }_{\Omega }{w}^{\alpha } = {\int }_{\Omega }{w}^{\alpha - 1}u - \delta {\int }_{\Omega }{w}^{\alpha } \]\n\n\[ \leq \frac{\delta }{2}{\int }_{\O...
Yes
Lemma 3.1 Let \( \delta > 0 \) and \( \alpha \geq 2 \) . Assume that\n\n\[ \mu > {\left( 2\epsilon \right) }^{\frac{1}{\alpha - 1}},\;0 < \epsilon < {2\delta }.\]\n\nThen the global solution of (3.1) has the property that there exists a constant \( \eta > 0 \) such that the nonnegative functions \( E \) and \( F \) def...
Proof The nonnegative of \( E\left( t\right) \) are similar to the recent works \( {}^{\left\lbrack {10},{16} - {17}\right\rbrack } \), so we omit it. In accordance with strong maximal principle and \( {U}_{0} > 0 \), it is readily yielded that \( U \) is positive in \( \bar{\Omega } \times \left( {0,\infty }\right) \)...
Yes
Theorem 2.1 Let \( \Phi \in \mathcal{G},0 < q < \infty \) and \( \omega \) be a nonnegative random variable on \( \left( {\Omega ,\mathcal{F}, P}\right) \) . If the martingale \( f = {\left( {f}_{n}\right) }_{n \geq 0} \) is in \( {H}_{\Phi, q,\omega }^{\sigma } \), then there exists a sequence \( {\left( {a}^{k}\right...
Proof Assume \( f \in {H}_{\Phi, q,\omega }^{\sigma } \) . Let us consider the stopping times for all \( k \in \mathbb{Z} \) , \n\n\[ \n{\tau }_{k} = \inf \left\{ {n \in \mathbb{N} : {\sigma }_{n + 1}\left( f\right) > {2}^{k}}\right\} ,\;\inf \varnothing = \infty . \n\] \n\nFor each stopping time \( {\tau }_{k} \), we ...
Yes
Theorem 2.2 Let \( \Phi \in \mathcal{G},0 < q < \infty \) and \( \omega \) be a nonnegative random variable on \( \left( {\Omega ,\mathcal{F}, P}\right) \) . If the martingale \( f = {\left( {f}_{n}\right) }_{n \geq 0} \) is in \( {\mathcal{Q}}_{\Phi, q,\omega } \), then there exists a sequence \( {\left( {a}^{k}\right...
Proof The proof is similar to the one of Theorem 2.1, so we only give it in sketch. If \( f = {\left( {f}_{n}\right) }_{n \geq 0} \in {\mathcal{Q}}_{\Phi, q,\omega } \), the stopping times \( {\tau }_{k} \) are defined by\n\n\[ \n{\tau }_{k} = \inf \left\{ {n \in \mathbb{N} : {\lambda }_{n} > {2}^{k}}\right\} ,\;\inf \...
Yes
Theorem 2.3 Let \( \Phi \in \mathcal{G},0 < q < \infty \) and \( \omega \) be a nonnegative random variable on \( \left( {\Omega ,\mathcal{F}, P}\right) \) . If the martingale \( f = {\left( {f}_{n}\right) }_{n \geq 0} \) is in \( {\mathcal{D}}_{\Phi, q,\omega } \), then there exists a sequence \( {\left( {a}^{k}\right...
Proof The only difference in the proof is that \( {\lambda }_{n} = \mathop{\sum }\limits_{{k \in \mathbb{Z}}}{\mu }_{k}{\begin{Vmatrix}S\left( {a}^{k}\right) \end{Vmatrix}}_{\infty }{\chi }_{\left\{ {\tau }_{k} \leq n\right\} } \) is replaced by \( {\lambda }_{n} = \mathop{\sum }\limits_{{k \in \mathbb{Z}}}{\mu }_{k}{\...
No
Lemma 4.1 Let \( \Phi \in \mathcal{G} \cap \Delta \cap \nabla ,0 < {r}_{1},{r}_{2}, p, q \leq \infty \) such that \( \frac{1}{r} = \frac{1}{{r}_{1}} + \frac{1}{{r}_{2}} \) . If \( f \in {L}_{{\Phi }^{\left( {r}_{1}\right) }}, g \in {L}_{{\Phi }^{\left( {r}_{2}\right) }} \), then \( {fg} \in {L}_{{\Phi }^{\left( r\right...
The proof is similar to Page 64 in [7] and we omit it.
No
Theorem 4.1 Let \( \Phi \in \mathcal{G} \cap \Delta \cap \nabla ,0 < {q}_{0},{q}_{1} < 1,0 < {r}_{1},{r}_{2} \leq \infty ,\omega \) is nonnegative random variable and\n\n\[ \n\frac{1}{r} = \frac{1}{{r}_{1}} + \frac{1}{{r}_{2}},\;\frac{1}{q} = \frac{1 - \theta }{{q}_{0}} + \frac{\theta }{{q}_{1}},0 < \theta < 1.\n\]\n\n...
Proof Let \( f \in {H}_{{\Phi }^{\left( {r}_{2}\right) }, q,\omega }^{\sigma } \) . There exists a sequence \( {\left( {a}^{k}\right) }_{k \in \mathbb{Z}} \) of \( \left( {1,\Phi ,\infty ,\omega }\right) \) atoms and a sequence \( {\left( {\mu }_{k}\right) }_{k \in \mathbb{Z}} \in {\ell }_{q} \) of real numbers such th...
Yes
Lemma 2.3 Let \( L \) be defined by (2.2), then\n\n\[ \operatorname{Ker}L = \left\{ {u \in X \mid u\left( t\right) = c,\forall t \in \left\lbrack {0, T}\right\rbrack, c \in {\mathbb{R}}^{2}}\right\} ,\n\]\n\n\[ \operatorname{Im}L = \left\{ {y \in Y \mid \left( \begin{matrix} {\int }_{0}^{T}{\left( T - s\right) }^{\alph...
Proof Obviously, by the Caputo fractional derivative and Riemann-Liouville fractional integral, we can see that (2.4) holds.\n\nIf \( y \in \operatorname{Im}L \), then there exists \( u \in \operatorname{dom}L \) such that \( y = {Lu} \) . That is, \( {y}_{1}\left( t\right) = {D}_{{0}^{ + }}^{\alpha }{u}_{1}\left( t\ri...
Yes
Lemma 2.4 Let \( L \) be defined by (2.2), then \( L \) is a Fredholm operator of index zero, and the linear continuous projector operators \( P : X \rightarrow X \) and \( Q : Y \rightarrow Y \) can be defined as\n\n\[ \n{Pu}\left( t\right) = \left( \begin{matrix} {t}^{1 - \alpha }{u}_{1}\left( t\right) { \mid }_{t = ...
Proof Obviously, \( \operatorname{Im}P = \operatorname{Ker}L \) and \( {P}^{2}u = {Pu} \) . It follows from \( u = \left( {u - {Pu}}\right) + {Pu} \) that \( X = \operatorname{Ker}P + \operatorname{Ker}L \) . By simple calculation, we can get that \( \operatorname{Ker}P \cap \operatorname{Ker}L = \{ 0\} \) . Then we ge...
Yes
Theorem 3.2 Suppose that the condition \( \left( {\mathrm{H}}_{2}\right) \) holds. Further, assume that the following condition holds.\n\n\( \left( {\mathrm{H}}_{3}\right) \) there exists a nonnegative number \( r \geq 0 \) such that\n\n\[ \mathop{\lim }\limits_{{\left| u\right| + \left| v\right| \rightarrow + \infty }...
Proof Note that \( {\left\lbrack \frac{4{T}^{\beta }r}{\Gamma \left( {\beta + 1}\right) }\left( 1 + \frac{{4}^{\nu - 1}{T}^{\alpha + \nu - 2}}{{\left( \Gamma \left( \alpha + 1\right) \right) }^{\nu - 1}}\right) \right\rbrack }^{\frac{1}{{P}_{m} - 1}} < 1 \), then there is a constant \( \varepsilon > 0 \) such that \( {...
Yes
Consider the following mixed fractional periodic boundary value problem with \( p\left( t\right) \) -Laplacian operator:
By Theorem 3.1, the problem (3.11) has at least one solution.
No
Lemma 2.6 The functions \( {g}_{1} \) and \( {g}_{2} \) given by (2.10) have the properties:\n\n(i) \( {g}_{1}\left( {t, s}\right) \leq \frac{B}{\Gamma \left( \alpha \right) }{t}^{\alpha - 1} \) for all \( t, s \in \left\lbrack {0,1}\right\rbrack \) ;\n\n(ii) \( {g}_{2}\left( {t, s}\right) \leq \frac{{t}^{\alpha - q - ...
Proof (i) From (2.10), we have\n\n\[ {g}_{1}\left( {t, s}\right) \leq \frac{1}{\Gamma \left( \alpha \right) }{t}^{\alpha - 1}\left\lbrack {B{\left( 1 - s\right) }^{\alpha - p - 1} - \left( {B - 1}\right) {\left( 1 - s\right) }^{\alpha - q}}\right\rbrack \]\n\n\[ \leq \frac{1}{\Gamma \left( \alpha \right) }{t}^{\alpha -...
Yes
Lemma 2.7 \( G\left( {t, s}\right) \) is continuous on \( \left\lbrack {0,1}\right\rbrack \times \left\lbrack {0,1}\right\rbrack \), and satisfies\n\n(i) \( G\left( {t, s}\right) \leq D{t}^{\alpha - 1} \) for all \( t, s \in \left\lbrack {0,1}\right\rbrack \), where \( D \) is given by (2.5);\n\n(ii) \( G\left( {t, s}\...
Proof By definition of the function \( G\left( {t, s}\right) \), we know that for all \( t, s \in \left\lbrack {0,1}\right\rbrack \), the function \( G\left( {t, s}\right) \) is continuous.\n\n(i) By Lemma 2.6, we obtain for all \( t, s \in \left\lbrack {0,1}\right\rbrack \),\n\n\[ G\left( {t, s}\right) = {g}_{1}\left(...
Yes
Theorem 3.1 \( T : P \rightarrow P \) is completely continuous.
Proof \( T : P \rightarrow P \) is continuous in view of nonnegative and continuity of \( G\left( {t, s}\right) \) and \( f \) . Next, we shall show that \( T \) is compact in \( P \) . Suppose \( \Omega \subset P \) is bounded, then, \( \exists M > 0 \) , \( \forall x \in \Omega \), s.t. \( \parallel x\parallel \leq M...
Yes