Advertisements

Mathematics Project Topics

Weak and Strong Convergence of an Iterative Algorithm for Lipschitz Pseudo-contractive Maps in Hilbert Spaces

Weak and Strong Convergence of an Iterative Algorithm for Lipschitz Pseudo-contractive Maps in Hilbert Spaces

Advertisements

Weak and Strong Convergence of an Iterative Algorithm for Lipschitz Pseudo-contractive Maps in Hilbert Spaces

Chapter One

PREAMBLE TO THE STUDY

Theย contributionย ofย thisย thesisย fallsย underย aย branchย ofย mathematicsย calledย Functionalย Analysis.ย Functionalย Analysisย asย anย independentย mathematicalย disciplineย startedย atย theย turnย ofย theย 19thย centuryย andย wasย finallyย establishedย inย 1920โ€™sย andย 1930โ€™s,ย onย oneย handย underย theย influenceย ofย theย studyย ofย specificย classesย ofย linearย operators-integralย operatorsย and integral equations connected with them-and on the other hand under the influence ofย the purely intrinsic development of modern mathematics with its desire to generalize andย thusย toย clarifyย theย trueย natureย ofย someย regularย behaviour.ย Quantumย Mechanicsย alsoย hadย a great influence on the development of Functional Analysis, since its basic concepts, forย example energy, turned out to be linear operators (which physicists at first rather looselyย interpreted as infinite dimensional matrices) on infinite dimensional spaces.

Theย contributionย ofย thisย thesisย fallsย underย aย branchย ofย mathematicsย calledย Functionalย Analysis.ย Functionalย Analysisย asย anย independentย mathematicalย disciplineย startedย atย theย turnย ofย theย 19thย centuryย andย wasย finallyย establishedย inย 1920โ€™sย andย 1930โ€™s,ย onย oneย handย underย theย influenceย ofย theย studyย ofย specificย classesย ofย linearย operators-integralย operatorsย and integral equations connected with them-and on the other hand under the influence ofย the purely intrinsic development of modern mathematics with its desire to generalize andย thusย toย clarifyย theย trueย natureย ofย someย regularย behaviour.ย Quantumย Mechanicsย alsoย hadย a great influence on the development of Functional Analysis, since its basic concepts, forย example energy, turned out to be linear operators (which physicists at first rather looselyย interpreted as infinite dimensional matrices) on infinite dimensional spaces.

Chapter Two

ย Preliminaries

Inย thisย Chapterย weย presentย mostlyย geometricย conditionsย ensuringย thatย convergenceย isย strong.ย Most of the properties will be established in the frame work of Hilbert spaces.ย Although a lot of results can be extended to larger classes of spaces, we will only do so inย someย specificย cases,ย sinceย ourย aimย isย toย emphasizeย unityย inย termsย ofย toolsย andย approach.

Definitions andย Technicalย Resultsย Aboutย Convergentย Sequencesย ofย Realย 

Definition (Strongย Convergence)

Letย Hย beย aย Hilbertย space.ย ย Weย sayย thatย {xn}ย convergesย (strongly)ย toย xย ifย limnโ†’โˆžย วxnย โˆ’xวย =

  1. Thisisย writtenย limnโ†’โˆžย xnย =ย 0ย orย simplyย xnย โ†’ย x.

Definition (Weakย Convergence)

Letย ย Hย ย beย ย aย ย Hilbert ย space.ย ย ย Weย ย sayย ย thatย ย theย ย sequenceย ย {xn}โˆžn=0ย ย ย ofย ย elementsย ย ofย ย aย ย Hilbert space Hย converges weakly to x โˆˆ Hย if there is an x โˆˆ Hย such that for every fย โˆˆ Hโˆ—,ย limnโ†’โˆžย fย (xn)ย ย =ย ย fย (x).ย ย ย Weย ย callย ย theย ย pointย ย xย ย aย ย weak ย limitย ย ofย ย theย ย sequencesย ย {xn}โˆžn=0ย ย ย and writeย xnย ~ย x.

Propositionย 2.1.3ย Aย sequenceย {xn}ย inย aย realย Hilbertย spaceย Hย convergesย weaklyย toย aย point

xย โˆˆย Hย ย ifย andย onlyย ifย limโŸจxn,ย zโŸฉย =ย โŸจx,ย zโŸฉย ย โˆ€ย zย โˆˆย H.

Advertisements

Proof:

Supposeย xnย ~ย x,ย thenย fย (xn)ย โ†’ย fย (x) ย ย ย โˆ€ย f ย โˆˆย Hโˆ—. ย Letย zย โˆˆย H ย beย arbitraryย (butย fixed).

Thenย defineย gย :ย Hย ย โ†’ย fย ย =ย เฆฉย orย ย ย Cย ย byย g(x)ย =ย โŸจx,ย zโŸฉ.ย ย Thenย gย โˆˆย Hโˆ—,ย since

g(ฮฑxย +ย ฮฒy)ย =ย โŸจฮฑxย +ย ฮฒy,ย zโŸฉย =ย โŸจฮฑx,ย zโŸฉย +ย โŸจฮฒy,ย zโŸฉย =ย ฮฑโŸจx,ย zโŸฉย +ย ฮฒโŸจy,ย zโŸฉย =ย ฮฑg(x)ย +ย ฮฒg(y) Also,ย |ย g(x)ย |=|ย โŸจx,ย zโŸฉย |โ‰คย วxววzว.ย ย Thisย impliesย thatย gย ย isย boundedย andย วgวย โ‰คย วzว.

Sinceย gย โˆˆย Hโˆ—ย andย xnย ~ย xย thenย g(xn)ย โ†’ย g(x)

โ‡’ย โŸจxn,ย zโŸฉย โ†’ย โŸจx,ย zโŸฉ

โ‡)ย Supposeย โŸจxn,ย zโŸฉย โ†’ย โŸจx,ย zโŸฉย โˆ€ย zย โˆˆย Hย ย thenย weย proveย thatย xnย ~ย x.

Letย fย โˆˆย Hโˆ—ย beย arbitrary.ย Thenย byย theย Riezย Representationย Theorem,ย thereย existsย aย uniqueย zย โˆˆย Hย ย suchย thatย fย (x)ย =ย โŸจx,ย zโŸฉย โˆ€ย xย โˆˆย H.ย ย Thusย fย (xn)ย =ย โŸจxn,ย zโŸฉย andย sinceย โŸจxn,ย zโŸฉย โ†’ย โŸจx,ย zโŸฉ, weย haveย thatย fย (xn)ย โ†’ย fย (x),ย whichย impliesย thatย xnย ~ย x

Definitionย 2.1.4ย Aย sequenceย {an}ย isย saidย toย beย monotoneย increasingย ifย an+1ย โ‰ฅย anย forย all

n โ‰ฅ 1, and monotone decreasing if an+1ย โ‰ค anย for n โ‰ฅ 1.ย We say {an} is monotone (orย monotonic)ย ifย itย isย ofย oneย ofย theseย twoย types.

Weย nowย stateย andย proveย theย followingย importantย resultsย concerningย sequencesย ofย realย numbers.

Lemmaย 2.1.1ย [19] Let {an} and {bn} be two sequences of a normed space Xย and {tn} aย sequenceย ofย realย numbers.ย Ifย theย followingย conditions

  • 0โ‰คย tnย โ‰คย tย <ย 1ย and โˆžn=1ย tnย =ย โˆž,
  • an+1=ย (1ย โˆ’ย tn)anย +ย tnbnย forย allย nย โ‰ฅย 1,

 

Chapter Three

Weakย andย Strongย Convergenceย ofย anย Iterativeย Algorithmย forย Lipschitzย Pseudo-Contractiveย Mapsย inย Hilbertย Spaces

ย ย Mainย Result

Inย Chapterย 2,ย weย presentedย mostlyย geometric,ย conditionsย ensuringย thatย theย convergenceย isย strong.

Inย thisย Chapter,ย weย introduceย aย modifiedย Ishikawaย iterationย forย Lipschitzย pseudocontractiveย mapย inย realย Hilbertย spaces.

Algorithmย 3.1.1ย Letย Hย beย aย realย Hilbertย spaceย andย Kย beย aย closedย convexย subsetย ofย H.

Letย ย Tย ย ย : ย Kย ย โ†’ย Kย ย beย ย aย ย Lipschitzย ย pseudo-contractiveย ย mappingย ย suchย ย thatย ย Fย (Tย ) โˆ….ย ย ย Let

{ฮฑn},ย {ฮฒn}ย andย {ฮณn}ย beย realย sequencesย inย (0,ย 1).ย Forย givenย x1ย โˆˆย K,ย letย {xn}ย beย generatedย iterativelyย by

xn+1ย =ย Pk[(1ย โˆ’ย ฮฑnย โˆ’ย ฮณn)xnย +ย ฮณnTyn];

ynย =ย (1ย โˆ’ย ฮฒn)xnย +ย ฮฒnTxn,ย nย โ‰ฅย 1 (3.0)

Theoremย 3.1.2ย Letย Hย beย aย realย Hilbertย spaceย andย Kย beย aย closedย convexย subsetย ofย H.ย Let

Tย : Kย โ†’ Kย beย aย L-Lipschitz ย pseudo-contractve ย mapping ย such ย that F (T )ย theย sequencesย {ฮฑn},ย {ฮณn},ย {ฮฒn}ย โˆˆย (0,ย 1)ย satisfying

(i)ย ฮฒn(1ย โˆ’ย ฮฑn)ย >ย ฮณnโˆ€nย โ‰ฅย 1ย (ii)ย limnโ†’โˆžย ฮฑnย =ย 0ย andย ฮฃฮฑnย =ย โˆž

โˆ….ย ย ย Assume

0<ย ฮฑย โ‰คย ฮณnย โ‰คย ฮฒnย โ‰คย ฮฒย <ย [

ย ย ย ย ย 1ย 1+L2+1]

forย allย nย โ‰ฅย 1.ย Thenย theย sequenceย {xn}ย generated

byย (3.0)ย stronglyย convergesย toย aย fixedย pointย ofย Tย .

Proofย ofย Theoremย 3.1.2

Sinceย Fย (Tย )ย /=ย โˆ…,ย weย canย takeย pย โˆˆย Fย (Tย ).ย ย Fromย (3.0)ย weย have

วxn+1ย โˆ’ย pวย ย = วPk[(1ย โˆ’ย ฮฑnย โˆ’ย ฮณn)xnย +ย ฮณnTyn]ย โˆ’ย pว

โ‰คย ย ว(1ย โˆ’ย ฮฑnย โˆ’ย ฮณn)xnย +ย ฮณnTynย โˆ’ย pว

= ว(1ย โˆ’ย ฮฑnย โˆ’ย ฮณn)(xnย โˆ’ย p)ย +ย ฮณn(Tynย โˆ’ย p)ย โˆ’ย ฮฑnpว

โ‰คย ย ว(1ย โˆ’ย ฮฑnย โˆ’ย ฮณn)(xnย โˆ’ย p)ย +ย ฮณn(Tynย โˆ’ย p)ว

+ฮฑnวpว (3.1)

Nowย consider

ว(1ย โˆ’ย ฮฑnย โˆ’ย ฮณn)(xnย โˆ’ย p)ย +ย ฮณn(Tynย โˆ’ย p)วย ย =ย ย ย ย ว(1ย โˆ’ย ฮฑn)[(1ย โˆ’ย ฮณn)(xnย โˆ’ย p)ย +ย ฮณn(Tynย โˆ’ย p)]

+ฮฑn[โˆ’ฮณnxnย +ย ฮณnTyn]ว2

References

  • I. Alber, โ€Metric and Generarized Projection Operators in Banach Spaces : prop-erties and applications,โ€ย in Theory and Applications of Non linear operators ofย Accretive and Monotone Type, vol.178 of Lecture Notes in Pure and Applied Math-ย ematics,ย pp.ย 15-50,ย Marcelย Dekker,ย Newย York,ย NY,ย USA,1996.
  • Asplund, Positivity of Duality Mappings, Bull.Amer.ย Math.ย Soc., 73 (1967),ย 200-203
  • Banach,ย Surย lesย Operationsย Dansย Lesย Emsemblesย Abstraitsย etย Leurย Applicationย Auxย Equationsย Integrals.ย Fundamentaย Mathematicaeย 3,ย 133-181ย (1922).
  • H. Bauschke, J. Borwein, On Projection Algorithms for Solving Convex Feasibility Problems. SIAM Rev. 38, 367-426 (1996).
  • Berinde,ย Generalizedย Contractionsย andย Applicationsย (Romanian),ย Edituraย cubย pressย 22,ย Baiamare,ย 1997.
  • Berinde,ย Iterativeย Approximationย ofย Fixedย Pointsย (Springer,ย Berlin,ย 2007).
  • Borweinย andย J.M.ย Bowein,ย Fixedย Pointย Iterationsย forย Realย Functions,ย J.Math.Anal.Appl.ย 157(1991),ย 112-126.
  • E.Browder,ย Fixedย Pointย Theoremsย forย Non-compactย Mappingsย inย Hilbertย Space.ย Proc.ย Nat.ย Acad.ย Sci.ย USAย 53,ย 1272-1276ย (1965).

Advertisements

WeCreativez WhatsApp Support
Our customer support team is here to answer your questions. Ask us anything!