An Enhanced Deep Reinforcement Learning-Based Global Router for VLSI Design

Figure 4

A reward line chart of two initialization methods.