Thus far, we have focused ourselves on canonical optimization problems where each decision variable is non-negative, and all bounds are inequalities. However, what would happen if we relaxed, or strengthened these conditions?
In this chapter, we explore noncanonical optimization problems. In Section 3.1 we consider the possibility of potentially negative decision variables, and explore the algebraic and geometric interpretations of this phonomena. In Section 3.2 we do the same for bounds defined by equality conditions, rather than inequalities. In Section 3.3 we show how to encode and solve non-canonical linear optimization using Sage.
