报告人： 王舸（北京宽客进化科技有限公司）
时间：2022-09-15 15:10-17:00
地点：Room 217, Guanghua Building 2
Abstract:
Generative artificial intelligence (AI) has begun to attract huge public interest recently. The MIT Technology review ranked generative AI as one of the most promising advances in the world of AI in the past decade. Generative AI models can learn the underlying patterns and the latent structure from the input real world data, and then use that to generate (or create) similar content and scenario. In the talk, we will introduce the various AI-based generative techniques such as generative adversarial network (GAN), diffusion model and variational autoencoder. As an emerging technology, it is set to bring many benefits to important industries, such as finance. We developed a time series generator of synthetic financial market data which can replicate the joint probability distribution of the original market data. The generator is then used to test the risk characteristics of a trading book or investment portfolios in the various scenarios and the worse cases, monitor the market risk in the event, analyse the hedging strategies, develop trading strategies using previously unseen data, or simulate high frequency trading activity in a limit order book. Our practices in the business confirm that our proposed framework consistently and significantly outperforms state-of-the-art benchmarks with respect to the measures of effectiveness and innovativeness.
About the Speaker:
Ge Wang, PhD, graduated from University of Texas at Austin. He then worked in Princeton Plasma Physics Lab, University of Texas at Austin and University of Wisconsin-Madison to study the nuclear fusion models and simulations on the supercomputer. Dr. Wang contributed tens of papers to various research topics across disciplines: AI for fusion science, the generative simulations and signal network, etc. He now is the chief scientist of a startup (北京宽客进化科技有限公司) in China. Their startup is developing and providing the advanced products and services based on the generative AI and data-centric AI for the banks and securities.
Place: Room 217, Guanghua Building 2
Your participation is warmly welcomed!
欢迎扫码关注北大统计科学中心公众号，了解更多讲座信息!
