Human-like Visual Learning and Reasoning

Peng Cui,Wenwu Zhu

DOI: https://doi.org/10.1145/3123266.3130144

2017-01-01

Abstract:Compared with machines, people are far better learners as they are capable of learning models(1) from very limited samples of a new category and make accurate predictions and reasoning accordingly. To learn from human learning process and investigate human-like visual learning and reasoning mechanism is a promising way to develop new effective learning methods. In this tutorial, several basic problems in human-like visual learning and reasoning will be discussed, including why we need to learn from human learning process, what are the research goals of human-like visual learning and reasoning, and how to embed human learning characteristics into learning models.