Department of Computer Science | Institute of Theoretical Computer Science
Page | Location | Description |
14 | Line -6 | The word transfered should be transferred |
17 | Paragraph containg third equation | The first sentence of the paragraph should start as follows: "Among all nonnegative solutions of these equations and inequalities..." |
29 | Line 10 of second paragraph | The number 452.5 should be 452.25 |
75 | First sentence after first LP | The sentence should start as "We want to stress that..." |
91 | Line 2 | y^{T}b≥0 should be y^{T}b≥0 (non-bold zero) |
96 | Paragraph after Lemma 6.5.3 | The set {tx: x ∈ D} should be {tx: x∈ D, t ≥ 0} |
104 | Line before equation | The words "an m x m matrix M with all entries nonnegative" should be replaced by "a nonnegative matrix M with m columns". |
111 | The first two equations | In the first equation, the fraction Qs_{k} / √s_{k}^{T} Q s_{k} should be replaced by Q_{k}a_{i} / √a_{i}^{T} Q_{k} a_{i}. In the second equation, the fraction Qs_{k}s_{k}^{T}Q / s_{k}^{T}Qs_{k} should accordingly be replaced by Q_{k}a_{i}a_{i}^{T}Q_{k} / a_{i}^{T}Q_{k}a_{i}. |
112 | First sentence of last paragraph | The sentence should start as "If, for example, all the a_{ij} and b_{i} are integers..." |
115, 116 | Line 2 of 7.2, Line 2 of p.116 | Narendra Karmarkar's last name is misspelled as "Karmakar". Also, Karmarkar was a researcher at AT & T Bell Laboratories when he invented his algorithm, not at IBM. |
120 | End of first paragraph | The conditions that are given on f and the g_{i} are in general not sufficient for the existence of Lagrange multipliers. The paragraph should continue as follows: "But the problem max x: x^{2}=0 shows that in general, additional conditions on x are needed. A sufficient condition is that the vectors ∇g_{1}(x),∇g_{2}(x),...,∇g_{m}(x) are linearly independent. However, if all the g_{i} are linear functions (as in our case), it is easy to see that this extra condition is redundant." |
126 | Last line | The word nontrivial should be removed |
127 | Equation defining M_{0} | In the third line of the matrix defining M_{0}, it should be b^{T} and c^{T} instead of b and c |
167 | Paragraph on Sparse solutions of underdetermined linear systems | In Line 4, the subordinate clause "the rows of A form a basis of its orthogonal complement" should read as "the rows of A lie in its orthogonal complement". Then the following sentence should be inserted: "In our application, A will have full rank and thus form a basis of the orthogonal complement". |
182 | Last line | In the inequality, the function ν* should be replaced by τ*. |
219 | 14th entry of right column | "Karmakar's algorithm" should be "Karmarkar's algorithm". |