notesum.ai

Published at December 4

MLQM: Machine Learning Approach for Accelerating Optimal Qubit Mapping

quant-ph

cs.ET

Released Date: December 4, 2024

Authors: Wenjie Sun¹, Xiaoyu Li¹, Lianhui Yu¹, Zhigang Wang¹, Geng Chen¹, Guowu Yang¹

Aff.: ¹University of Electronic Science and Technology of China

Arxiv: http://arxiv.org/pdf/2412.03249v1

Refer to caption

Samples	Time consumption (Seconds)		Memory (MB)		Search Counts
Samples	OLSQ2 (DAC’23)	MLQM (Ours)	OLSQ2 (DAC’23)	MLQM (Ours)	OLSQ2 (DAC’23)	MLQM (Ours)
wstate_n3	271	178 (1.52x)	628	606 ( $\downarrow$ 3.5%)	11	6 ( $\downarrow$ 45.5%)
bv_n14	605	458 (1.32x)	1591	1605 ( $\uparrow$ 0.9%)	17	8 ( $\downarrow$ 52.9%)
fredkin_n3	645	403 (1.6x)	574	494 ( $\downarrow$ 14.1%)	9	8 ( $\downarrow$ 11.1%)
ghz_state_n23	821	180 (4.56x)	1568	1157 ( $\downarrow$ 26.2%)	29	4 ( $\downarrow$ 86.2%)
simon_n6	1119	1580 (0.71x)	1447	1469 ( $\uparrow$ 1.5%)	11	7 ( $\downarrow$ 36.3%)
basis_test_n4	1199	885 (1.35x)	5369	3286 ( $\downarrow$ 38.8%))	6	4 ( $\downarrow$ 33.3%)
cat_n35	5774	851 (6.78x)	4018	2293 ( $\downarrow$ 42.9%)	40	4 ( $\downarrow$ 90.0%)
ghz_n40	15282	2706 (5.65x)	4615	3131 ( $\downarrow$ 32.2%)	44	6 ( $\downarrow$ 86.4%)