Nonlinear Programming Book, 2nd Edition
dimi...@mit.edu
NONLINEAR PROGRAMMING: 2nd EDITION
by Dimitri P. Bertsekas
ISBN: 1-886529-14-0
(780 pages, Hardcover, appeared September 1999)
This is a substantially expanded (by 130 pages) and improved edition of
the author's
1995 nonlinear programming book. The treatment focuses on iterative
algorithms for
constrained and unconstrained optimization, Lagrange multipliers and
duality, large
scale problems, and on the interface between continuous and discrete
optimization.
Nearly 40% of the new material represents miscellaneous additions
scattered throughout
the text. The remainder deals with three new topics. These are:
a) A new section in Chapter 3 that focuses on a particularly simple but
far-reaching treatment of Fritz John necessary conditions and constraint
qualifications, and also includes
semi-infinite programming.
b) A new section in Chapter 5 on the use of duality and Lagrangian
relaxation for
solving discrete optimization problems. This section describes several
motivating
applications, and provides a connecting link between continuous and
discrete
optimization.
c) A new section in Chapter 6 on approximate and incremental subgradient
methods. This
material is the subject of ongoing research, but it was thought
sufficiently significant
to be included in summary here.
A new internet-based feature was added to the book, which significantly
extends its
scope and coverage. Many of the theoretical exercises, quite a few of
them new, have
been solved in detail and their solutions have been posted in the book's
www page
http://world.std.com/~athenasc/nonlinbook.html
which also gives the preface and the detailed table of contents.
Among its special features, the book:
-----------------------------------------------------------------------
** provides extensive coverage and a unifying analytical framework for
iterative optimization methods
** provides a detailed treatment of interior point methods for linear
programming
** covers in depth duality theory from both a variational and a
geometrical/convex analysis point of view
** includes much new material on a number of topics, such as neural
network training, discrete-time optimal control, and large-scale
optimization
** includes a large number of examples and exercises. Has been
developed through extensive classroom use in first-year graduate
courses
-----------------------------------------------------------------------
Chapter Titles:
1. Unconstrained Optimization
2. Optimization Over a Convex Set
3. Lagrange Multiplier Theory
4. Lagrange Multiplier Algorithms
5. Duality and Convex Programming
6. Dual Methods
Appendixes
********************************************************************
Publisher's Information:
Athena Scientific, P.O.Box 391, Belmont, MA, 02178-9998, U.S.A.
Email: athe...@world.std.com
Tel: (617) 489-3097, FAX: (617) 489-2017
http://world.std.com/~athenasc/index.html
Nonlinear Programming ($79.00, 780 pp., Hardcover, ISBN: 1-886529-00-0)
********************************************************************
Tuomo Takkula
Hi
> **********************************************************************
>
> NONLINEAR PROGRAMMING: 2nd EDITION
> by Dimitri P. Bertsekas
> ISBN: 1-886529-14-0
> (780 pages, Hardcover, appeared September 1999)
>
> This is a substantially expanded (by 130 pages) and improved edition of
> the author's
> 1995 nonlinear programming book. The treatment focuses on iterative
> algorithms for
> constrained and unconstrained optimization, Lagrange multipliers and
> duality, large
> scale problems, and on the interface between continuous and discrete
> optimization.
>
> Nearly 40% of the new material represents miscellaneous additions
> scattered throughout
> the text. The remainder deals with three new topics. These are:
<snip>
Is there a possibility for a cheap "upgrade" from the first to the
second edition?
I bought the first edition of the book and although I found it very
nice, it contained an unusual (annoying) amount of mistakes.
So, a economy-upgrade for owners of the first edition would be
appropriate, wouldn't it?
Best regards
Tuomo Takkula