Exploring Machine Teaching for Object Recognition with the Crowd.
CHI Extended Abstracts(2019)
摘要
Teachable interfaces can enable end-users to personalize machine learning applications by explicitly providing a few training examples. They promise higher robustness in the real world by significantly constraining conditions of the learning task to a specific user and their environment. While facilitating user control, their effectiveness can be hindered by lack of expertise or misconceptions. Through a mobile teachable testbed in Amazon Mechanical Turk, we explore how non-experts conceptualize, experience, and reflect on their engagement with machine teaching in the context of object recognition.
更多查看译文
关键词
crowdsourcing, interactive machine learning, object recognition, teachable machines
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络