Another gem every problem solver or math olympic should have on the shelf. It encapsulates problem solving techniques in a handful of concepts, providing several examples and a whole bunch of problems (around 1300 I guess), most of them with solutions or advanced hints.
The book is divided into chapters consisting of a "strategy" or a subject. The range of problems is really wide, and sometimes are quite difficult. In my opinion there is a missing chapter on symmetry, for its applicability on problem solving.
"An experienced problem solver can often infer the road to the solution from the result." - grab all the information the problem gives you.
The chapters are:
Chapter 1 - The invariance principle
Here Engel describes invariants and monovariants. The idea is to construct a function ( often integer, positive) that does not change or is bounded and monotonic (in the latter case the algorithm must terminate)
- If you have (or can introduce a transformation), look for an invariant. - Distance from the origin is frequently used on these problems - parity is also useful - If there is repetition, look for what does not change.
Chapter 2 - Coloring proofs
Weird chapter on solving board problems. Typical colorings are: chess-like, column (or row) based, etc. A few parity problems on this chapter.
Chapter 3 - The extremal principle
Very wide-applicable strategy. Many examples on Number Theory, Geometry, Combinatorics, etc are given.
- Pick an object that maximizes of minimizes some function. It frequently has nice properties, allows to simplify or even to arrive in a contradiction. - Every finite nonempty set of nonnegative numbers has a min and a max - Every nonempty set of positive integers has a min ( well ordering principle)
Chapter 4 - The box principle ( pigeonhole )
- If n+1 pearls are put into n boxes, then at least one box has more than one pearl.
The difficulty on this chapter is to find the box and the pearls for each problem.
Chapter 5 - Enumerative Combinatorics
- Count by bijection - Recursion - Divide and conquer : split a problem into smaller parts, solve the small problem and combine solutions. - Sum rule, product rule, product-sum rule, sieving - Construct of a graph which accepts the objects to be counted. - Count in two ways - If you cannot find the number of good objects, find the number of bad objects.
Chapter 6 - Number Theory
Basic introduction to NT. Lots of problems.
Chapter 7 - Inequalities
Basic Inequalities ( CS , x2> 0, AM-GM-HM, rearrangement, chebyshev, etc). A few hints for solving such problems:
-does the expression remind you AM-GM-HM? Cauchy-Schwarz? -Can you apply rearrangement? -An inequality homogeneous on its variables can be normalized. -Is there any symmetry on the variables? In that case you can make additional hypothesis (e.g. a>= b >=c >=...) -Trigonometrical substitution? -Convexity and concavity ( Jensen) -Can you transform it into a simpler form?
Chapter 8 - The induction Principle
Small Chapter on induction.
Chapter 9 - Sequences
Nice chapter on sequences. Includes linear recurrence, josephus problem, etc.
Chapter 10 - Polynomials
- Vieta's theorem - Fundamental theorem of algebra - Roots of unity - reciprocal polynomials - Symmetric polynomials
Chapter 11 - Functional Equations
Cauchy's functional equation and others. Small chapter
Chapter 12 - Geometry
Very long chapter, including: - geometry and complex numbers - transformation geometry - classical euclidean geometry
Chapter 13 - Games
Describes nim, Bachet, Wythoff, etc
Chapter 14 - Further Strategies
- Graph theory - Infinite descent - Working backwards - Conjugate numbers - equations, functions and iterations - integer funcitons (floor, ceiling, etc)
I really liked this book when I read it in high school.It has numerous problems ranging from hard to routine and can be used effectively for training for mathematical contests and for problem solving in general.