Display Pixel Layout Design with Deep Reinforcement Learning
TimeTuesday, July 12th6pm - 7pm PDT
LocationLevel 2 Lobby
Event Type
Networking Reception
Work-in-Progress Poster
DescriptionDisplay pixel layout design is one of the most complex and time-consuming stages of display panel development. Previously, pixel layout design has been regarded as a highly specialized area of experts because of its complexity. However, with the evolution of AI, it can be applied even in layout design. In this work, we formulate the pixel layout design as a sequential optimization problem of reinforcement learning. The objective of this study is to design the pixel layout with a maximum aperture ratio by reinforcement learning agent. Through our proposed approach, we obtained a competitive pixel layout compared to the experts.