A study of DQN using VisionTransformer as an image extractor
Toshiki Hatano, Toi Tsuneda, Satoshi Yamane · 2021 IEEE 10th Global Conference on Consumer Electronics (GCCE) · 2021
In recent years, the field of image recognition has seen rapid technological innovation, and in addition to the CNN-based models used in the past, models using transformers have also been devised. However, there are no results on the verification of DQN using the latest image recognition models in existing research. In this study, we examined the performance of DQN using VisionTransformer as an image extractor, which has become a hot topic in recent years. The performance was not stable due to problems caused by the different architecture from the resulting CNN.