{"title":"紧线性算子","authors":"F. Bonsall","doi":"10.1017/9781139030267.016","DOIUrl":null,"url":null,"abstract":"Proof. Since the dimensions of R(A) is always small or equal to the dimension of N (A)⊥, X and R(A), and N (A)⊥ are all infinite dimensional. Hence, we can find a sequence (xn) with xn ∈ N (A)⊥ with ‖xn‖ = 1 and 〈xn, xm〉 = 0 for n 6= m. Since A is compact, the sequence (yn) = (Axn) has to contain a convergent subsequence. Thus, for any δ > 0 we can find k and l such that ‖yk − yl‖ < δ. However, ‖A(yk − yl)‖ = ‖xk − xl‖ = ‖xk‖ + ‖xl‖ − 2〈xk, xl〉 = 2, although A†0 = 0. Hence, A† is not continuous.","PeriodicalId":256579,"journal":{"name":"An Introduction to Functional Analysis","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Compact Linear Operators\",\"authors\":\"F. Bonsall\",\"doi\":\"10.1017/9781139030267.016\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Proof. Since the dimensions of R(A) is always small or equal to the dimension of N (A)⊥, X and R(A), and N (A)⊥ are all infinite dimensional. Hence, we can find a sequence (xn) with xn ∈ N (A)⊥ with ‖xn‖ = 1 and 〈xn, xm〉 = 0 for n 6= m. Since A is compact, the sequence (yn) = (Axn) has to contain a convergent subsequence. Thus, for any δ > 0 we can find k and l such that ‖yk − yl‖ < δ. However, ‖A(yk − yl)‖ = ‖xk − xl‖ = ‖xk‖ + ‖xl‖ − 2〈xk, xl〉 = 2, although A†0 = 0. Hence, A† is not continuous.\",\"PeriodicalId\":256579,\"journal\":{\"name\":\"An Introduction to Functional Analysis\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-03-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"An Introduction to Functional Analysis\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1017/9781139030267.016\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"An Introduction to Functional Analysis","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1017/9781139030267.016","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Proof. Since the dimensions of R(A) is always small or equal to the dimension of N (A)⊥, X and R(A), and N (A)⊥ are all infinite dimensional. Hence, we can find a sequence (xn) with xn ∈ N (A)⊥ with ‖xn‖ = 1 and 〈xn, xm〉 = 0 for n 6= m. Since A is compact, the sequence (yn) = (Axn) has to contain a convergent subsequence. Thus, for any δ > 0 we can find k and l such that ‖yk − yl‖ < δ. However, ‖A(yk − yl)‖ = ‖xk − xl‖ = ‖xk‖ + ‖xl‖ − 2〈xk, xl〉 = 2, although A†0 = 0. Hence, A† is not continuous.