Topic: Time Series Prediction Using Neural Networks With Attention
Speaker: Dr. Thomas Li, Ph.D. in Mathematics, Courant Institute

Time: 8/31/2025, Sunday 8:30 PM EDT

Abstract
Time series prediction is a popular research topic. Instead of utilising the original time series data, we can predict the future trends by utilising charts generated from the data. In this seminar, we will talk about the theoretic foundations about convolutional neural network and attention mechanisms, and the implementation for time series prediction utilising charts.

时间序列预测是一个热门的研究领域。与直接使用原始时间序列数据不同，我们可以通过利用由数据生成的图像来预测未来趋势。在本次研讨会中，我们将讨论卷积神经网络和注意力机制的理论基础，以及基于图像进行时间序列预测的实现方法。

Speaker bio
Dr. Thomas Li completed his Ph.D. in mathematics at the Courant Institute. His research interests include stochastic control and optimization, optimal transport, numerical analysis, and machine/deep learning. He brings extensive expertise in mathematical foundations of AI and their practical applications.