GeniusRoute: A New Analog Routing Paradigm Using Generative Neural Network Guidance

A customized graph neural network model for guiding analog IC placement

“…The value of FOM is in [0, 1] and ideally 1. This definition is consistent with the performance cost defined in Equations (17) and (18).…”

“…Five variants of performance driven placement based on Section 4 are tested. They are guided by combinations of PEA vs. CNN [19], SS (Self-sustained learning) vs. transfer learning, and performance cost defined by Equation (17) and defined by Equation (18). To capture the overall circuit performance, a…”

“…Recently, various machine learning techniques have been explored for analog circuit synthesis [15][16][17][18][19][20]. In [15], GNN (Graph Neural Network) is applied to produce layout templates for passive elements in RF circuits.…”

“…A GAN (Generative Adversarial Network)based well generation technique is proposed for analog circuit designs [16]. Variational auto-encoder is employed in [17] to learn from manual layout and provide routing guidance. The work of [18] makes use of GCN (Graph Convolutional Network) to recognize analog sub-circuits.…”

Re-examining VLSI manufacturing and yield through the lens of deep learning

“…Besides, the ML-inspired placement framework, DREAMPlace [29], has demonstrated unprecedented acceleration for the placement process. Moreover, ML has been used for routing guidance [30] and routing violation detection [28].…”