This project implements the efficient method for the target function approximation for stochastic circuit minimization proposed in our paper [1].
Reference paper(s):
-
[1]: Chen Wang, Weihua Xiao, John Hayes, and Weikang Qian, "Exploring target function approximation for stochastic circuit minimization," in Proceedings of the 2020 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), virtual event, San Diego, CA, USA, 2020, pp. 122:1-122:9.
-
[2]: Xuesong Peng and Weikang Qian, "Stochastic circuit synthesis by cube assignment," in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 37, no. 12, pp. 3109-3122, 2018.
The following two directories provide the two major programs.
-
Bernstein Polynomial Approximation by Matlab is supposed to be run before TFASC program to prepare the input file for it.
-
TFASC main program (on 64-bit Linux) is the program for the Dynamic Approximation (DA) method as the best method proposed in our paper [1].
-
pxs method (no approximation) is the program for stochastic circuit synthesis by cube assignment based on [2], which is treated as the baseline method in [1]. For this method, the initial feature vector (problem vector) is realized exactly with no change.
Please refer to readme.md in each directory for more details.
If you have any questions or suggestions, please feel free to eamil to wangchen_2011@sjtu.edu.cn, thanks!