Wolfram Computation Meets Knowledge

Wolfram Summer School


Yiqun Liu

Science and Technology

Class of 2016


Yiqun is a junior at Tianjin Nankai High School. He’s interested in exploring topics related to machine learning, such as natural language processing, pattern recognition and statistical learning. In his spare time, he likes to read notes and watch online courses related to these topics. His dream is to become a master in machine intelligence and make the world a better place. Of course, he appreciates the chance to attend this unique summer program.

Project: Neural Networks Learn Arithmetic Operations

With the continuous stream of innovations in machine intelligence, neural networks achieve remarkable success on a spectrum of tasks. Nonetheless, can machines learn to think at an abstract level? Will neural networks master mathematical logic under supervised learning? In this project, my goal is to explore the optimal neural network model for learning arithmetic operations like addition, multiplication and sorting.

Although learning algorithms from examples is feasible in a number of ways—memory networks, LSTM networks, neural Turing machines, etc.—current models are either too complicated to be trained efficiently or too restricted to generalize.

Favorite 3-Color 2D Totalistic Cellular Automaton

Rule 510510