genetic algorithms: do not work for most problems. develop solutions to particular problems using inheritance, crossover, and mutation. represent knowledge as groups of characteristics. are based on logic. seek to emulate a human expert's way of solving problems.