Department of

November 2017December 2017 January 2018 Su Mo Tu WeThFr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa 1 2 3 4 1 2 1 2 3 4 5 6 5 6 7 8 9 10 11 3 4 5 6 7 8 9 7 8 9 10 11 12 13 12 13 14 15 16 17 18 10 11 12 13 14 15 16 14 15 16 17 18 19 20 19 20 21 222324 25 17 18 19 20 21 22 23 21 22 23 24 25 26 27 26 27 28 29 30 24 25 26 27 28 29 30 28 29 30 31 31

Friday, December 1, 2017

**Abstract:** Machine learning has become an important technique to understand and predict the trends of large volume of data. While most machine learning models are static, static is hardly the case in real life. We need to create dynamical models by generalizing from past experience and results. In this talk, I will explore the usages of dynamical systems in machine learning. The talk will be divided into two parts. First, I will focus on the Kalman filter, a famous Bayesian model permitting exact inference in a discrete dynamical system, and its extensions. Then, I will discuss how to use continuous dynamical systems as a tool for machine learning, especially for deep neural networks. Some of the results are from my previous internship experience.