====================================================
The following is the updated version, including Pan, Xuanyu's and Zhu,
Yi's suggestion.
====================================================
(Zhu) I am sorry in the new list I did not include anything about
algebraic geoemtry. I think I will write a much more detailed book
list for learning algebraic geometry (on graduate level). For those
who are interested in algebraic geometry, I only have one
recommendation Joe Harris, Algebraic geometry: an introduction. Read
every chapters and do every exercises (I am not kidding! I am very
serious about this if you want to be an algebraic geometer). Plus, GSM
55, A Scrapbook of Complex Curve Theory, C. Clemens, is also a very
good book. Try to read it.
====================================================
DIFFERENTIAL GEOMETRY AND TOPOLOGY
(Zhu) Mathematics is a unified object. I dont want to put any book in
any smaller category since it is meaningless. To learn math, learn
everything.
***** means five star
1. Modern Geometry ***** ***** ***** (Strongly recommended to after
second year)
(Zhu) Read as much as possible. This three volume book is the best
book to learn differential topology and differential geometry. First
volumes are mainly about differential geometry in higher dimensional
case, including lots of useful examples. Second volume is the standard
beginning for the differential manifold. Volume 2 is the best book for
manifold theory. It is better than Warner's and Lee's. Volume 3 deals
with differential topology and algebraic topology on manifold,
including the Morse theory.
2. Lectures on Geometry vol. 1-6, Postnikov ****** (Strongly
recommended to first year)
(Zhu) 1st and 2nd have chinese versions in Yuquan's library. Those are
the pre-modern geometry books, best for the first year. We can see how
linear algebra working in higher dimensional geometry.
3. Topology, James R.Munkres **
(Zhu) Bad. Wong's point-set topology is better. You should go through
point-set topology asap. Dont spend too much time on this. It is a
language for pure math.
4. Prof Wong's notes *****
(Zhu) Good notes! You can read it very quickly and then go to
differential geometry to use this language. Again do not spend too
much time.
5. Differential Geometry of Curves and Surfaces, Manfredo P.do Carmo
*****
(Zhu) First course in differential geometry, do every exercises and
examples. Learn the computation, guys!! The computation is the key to
differential geometry.
6. Introduction to Smooth Manifolds, John M.Lee
(Zhu) BAD. Use Modern geometry, or any chinese book. Try to learn how
not compute the tensor is kind of more important than the concept
itself. Lee's book is a combination of everything, but a bad
combination. Learn computation in modern geometry!!!
7. Shen Yibing's book ***
(Zhu) First book in Riemannian geometry. Learn how to manipulate the
tensor stuff!
8. 黎曼几何初步, 黎曼几何选讲, 伍鸿熙 *****
(Zhu) Good! Read remarks every chapters before you read it seriously.
Get some idea beore you learn.
9. Morse Theory, J.Milnor *****
(Zhu) The Best book in Riemannian geometry and topology. Milnor can
always treat stuff in a concise and beautiful way. If you want to know
the topology of the manifold, start from the Morse theory. If you want
to get some sense in Riemannian geometry after learning the manifold
theory, Milnor's book has the best short introduction to Riemannian
geometry.
11. Topology from the Differentiable Viewpoint, J.Milnor
(Zhu) Good book but not good as Morse theory.
12. Algebraic Topology: An Introduction,William S.Massy *****
(Zhu) The best first course in algebraic topology forever..... Do
every exercises. Play some toy models.
13. Algebraic Topology,Allen Hatcher,
www.math.cornell.edu/~hatcher
*****
(Pan) Hatcher Chapter 0 to Chapter 4 (do most of the exercise). This
book is interesting, teaching you how to think problems by
drawing....But the book does not offer you a lot of power tools.....
(Zhu) The best second course in algebraic topology so far.... Do as
much exercises as you can. Play lots of toy models. I like its
treatment in homology theory, but not very much for cohomology theory.
The reason is that no one use singular cohomology to do geometry. It
should emphasize the relation with de Rham's cohomology and Cech's
cohomology. The best treatment is in Bott-Tu's book. Also, I hate the
proof for the Poincare duality in the book. The better proof can be
find in Bott-Tu. My favourate prooves are using Morse theory in Modern
geometry vol 3 and Griffiths-Harris Chapter 0.4.
14. Differential Forms in Algebraic topology, Bott-Tu ***** *****
***** (strongly recommended for every one)
(Pan) This book is very deep, however, very readable, even though you
are from physics department......You can touch the modern concepts of
topology
around 1940-1950s.....a lot of thing can be generalized to
algebraic
geometry case.... (Remark from Yi Zhu, You can generalize
every thing in Bott-Tu to algebraic geometry.)
(Zhu) The Best book in algebraic topology, differential topology,
differential geometry, algebraic geometry, complex geometry forever.
If you want to be a mathematician, put the book in your bed and read
it every night. Key words: First chern classes for line bundls, Leray
spectral sequences, Chern Classes. This book tells you how to do math
by "clever" computations. It provides you geometric intuitions of
cohomology theory and its huge applications. You will see those ideas
transplanted into algebraic/complex geometry by Grothendieck, and
become much more essential than the algebraic topology itself.
15. Characteristic Class, Milnor *****
(Pan) It is too good to say nothing.....Its content is an apex of
1940s' topology!!!
16. Topology of Fibre Bundle, Steenord *****
(Pan) You can learn obstruction theory from this book, read it is
enjoyable because Steenord tell you the pictures behind every
theorem...if u know how to construct the "topological moduli" space
(just infinity projective space), u will appreciate algebraic
stack.......
17. K-theory, Atiyah *****
(Pan) It is about thoery of vector bundle.....It is preparation for
Atiyah-
Singer Index theorem....If u are familiar with chapter 4 in
Hatcher, that the only difficult part of K-theory is Bott
periodicity.....Milnor's Morse theory prove it, but Atiyah only use
linear algbra, amazing......The algebraic K-theory transform a lot of
idea from topological one......As usual, Atiyah's book is very
elegant....
18. A Concise Course for Algebraic Topology, J.May ****
(Pan) It is good reference and u can read it if u have the background
of
previous books......It establish homotopy theory and homology
theory
in a modern view which can be generalized to algebraic geometry
( like
A-homotopy, model Category)....
(Zhu) To me, the book is too formal. I am willing to learn this stuff
only if I get the motivation.
===========================================================================
=====
ANALYSIS
1. Principles of Mathematical Analysis, W. Rudin *****
(Zhu) The Best Second course in calculus. But for the first course, I
still recommend Zorich, Mathematical analysis and 菲赫今哥尔茨 "微积分学教程". The
latter is still my favorite.
2. Complex Analysis, Ahlfors *****
(Zhu) I really like this book although it is not possible to finish
in
one semester. Gong Sheng's complex analysis book is much more
concise.
3. Ordinary Differential Eqations, Arnold
4. 数学物理方程,谷超豪,李大潜 高等教育出版社 *****
(Zhu) I really like this book. The book taught me that to learn
analysis, learn PDE first and to learn PDE, learn physics first. I am
still amazed by the magic idea of Fourier.
5. Stein's three analysis books ***** ***** *****
(Zhu) The Best analysis books for undergraduates!
===============================================
ALGEBRAIC STUFF.
1. Algebra, Michael Artin *****
(Zhu) No doubt it is the best. Do every exercises!
2. Introduction to Commutative algebra, Atiyah, Macdonald ******
(Zhu) Strongly recommended for those interested in algebraic geometry
and number theory. Do every exercises!
3. An Introduction to Homological Algebra, Charles A.Weibel
4. A course in Homological Algebra, P. Hilton; U.Stammbach. GTM 4
(Zhu) No comments. Homological algebra is purely geometric to me. You
should learn algebraic topology, Bott-Tu, Hartshorne rather than
these
books.
===============================================
Number Theory
1. Serre's A Course in Arithmetic ************
(Zhu) Read it! Get some feeling for number theory.
2. Numer theory 1 Fermat's Dream *************
(Zhu) This is written by 3 Japanese number theorist. I like it very
much. The authors tells you what is going on in number theory. Lots of
deep ideas is explained, like quadratic reciprocity law and rational
points on conics, Mordell's theorem and conjeture, Kummer's trick on
Last Fermat conjecture, Zeta functions. No chinese edition yet. I will
mail the book to Prof. Wong. I hope you can see it soon.
===============================================
It will be better if you can try to understand some basic PHYSICS.
1. 经典力学的数学方法 Arnold
2. 费曼物理学讲义1 2 3 Feynman
(Zhu) Keep your eyes on it. I cannot describe it properly.....