Approximating the Permanent with Deep Rejection Sampling
About
We present a randomized approximation scheme for the permanent of a matrix with nonnegative entries. Our scheme extends a recursive rejection sampling method of Huber and Law (SODA 2008) by replacing the upper bound for the permanent with a linear combination of the subproblem bounds at a moderately large depth of the recursion tree. This method, we call deep rejection sampling, is empirically shown to outperform the basic, depth-zero variant, as well as a related method by Kuck et al. (NeurIPS 2019). We analyze the expected running time of the scheme on random $(0, 1)$-matrices where each entry is independently $1$ with probability $p$. Our bound is superior to a previous one for $p$ less than $1/5$, matching another bound that was known to hold when every row and column has density exactly $p$.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Permanent approximation | Network Repository ENZYMES-g192 | Expected Running Time (s)10 | 8 | |
| Permanent approximation | Network Repository ENZYMES-g230 | Expected Running Time (s)10 | 8 | |
| Permanent approximation | Network Repository ENZYMES-g479 | Expected Running Time (s)7 | 8 | |
| Permanent approximation | Network Repository bcspwr01 | Running Time (s)10 | 8 | |
| Permanent approximation | Network Repository cage5 | Expected running time (s)10 | 7 | |
| Permanent approximation | Staircase-30 | Expected Running Time (s)7 | 6 | |
| Permanent approximation | Staircase-45 | Expected Running Time (s)800 | 2 |