1. [Publications](/publications)
2. Efficient Arithmetic Block Identification with Graph Learning and Network-flow
 
 # Efficient Arithmetic Block Identification with Graph Learning and Network-flow

  ![](/sites/default/files/styles/wide/public/publications/Weixin%20Screenshot_20230606115726_0.png?itok=EptKecjp)

 Arithmetic block identification in gate-level netlists plays an essential role for various purposes, including malicious logic detection, functional verification, or macro-block optimization. However, current methods usually suffer from either low performance or poor scalability. To address the issue, we come up with a novel framework based on graph learning and network flow analysis, that extracts desired logic components from a complete circuit netlist. We design a novel asynchronous bidirectional graph neural network (ABGNN) dedicated to representation learning on directed acyclic graphs. In addition, we develop a convex cost network-flow-based datapath extraction approach to match the predicted block inputs with predicted block outputs. Experimental results on open-source RISC-V CPU designs demonstrate that our proposed solution significantly outperforms several state-of-the-art arithmetic block identification flows.



 ## Authors



Ziyi Wang (CUHK)

Zhuolun He (CUHK)

Chen Bai (CUHK)

[Haoyu Yang](/person/haoyu-yang)

Bei Yu (CUHK)

 

 

 ## Publication Date



Thursday, December 8, 2022

 

 ## Published in



[ IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems ](https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9975800)

 

 ## Research Area



[Artificial Intelligence and Machine Learning ](/research-area/machine-learning-artificial-intelligence)

[Circuits and VLSI Design](/research-area/circuits)

 

 

 ## Uploaded Files



[Efficient\_Arithmetic\_Block\_Identification\_with\_Graph\_Learning\_and\_Network-flow.pdf](https://d1qx31qr3h6wln.cloudfront.net/publications/Efficient_Arithmetic_Block_Identification_with_Graph_Learning_and_Network-flow.pdf "Open file in new window")5.92 MB