arXiv++ Combinatorics

We consider optimization of nonlinear objective functions that balance $d$ linear criteria over $n$-element independence systems presented by linear-optimization oracles. For $d=1$, we have previously shown that an $r$-best approximate solution can be found in polynomial time. Here, using an extended Erdős-Ko-Rado theorem of Frankl, we show that for $d=2$, finding a $ρn$-best solution requires exponential time.