Seminar on Artificial Intelligence

Abstract

This seminar discusses the cutting edge of artificial intelligence today. First, regarding large language models, we discussed the principles of Attention, the use of Huggingface, and subsequent theoretical improvements and architectural innovations. For reinforcement learning, we started with classical reinforcement learning and discussed algorithms such as DQN and PG along with their various applications. Concerning generative models, we explored their development from VAE to the recently popular Stable Diffusion.

Time and Location

Time: 2:00pm-4:00pm, Tuesday, 2025 Spring

Location: Online, Tencent Meetings

Learning Materials

Materials are temporarily stored on Ruike Cloud Drive and Github: PPT, video material.