Application Research of Teaching Robot Based on ChatGPT

Abstract

The release of the artificial intelligence product ChatGPT has triggered a wave of attention. ChatGPT combines SMOTE with a generative adversarial network to establish a teaching robot OSW model for preprocessing training samples. Use Wasserstein GANN to study the distribution of small classes of data to build a balanced teaching robot system. Using ChatGPT to build TBD's Transformer and dual-channel long-short-term memory-deep network. This project intends to use the teaching machine learning algorithm based on BiLSTM to effectively acquire the long sequence features of the existing data. The research shows that the teaching robot based on ChatGPT proposed in this paper has higher teaching efficiency.