Machine Teaching

Machine teaching is the optimal control of machine learning. A machine learning algorithm defines a dynamical system where the state (learned model) is driven by training data. Machine teaching finds the optimal training data to drive the learning algorithm to a target model.

Machine teaching has applications in adversarial learning and education. In adversarial learning the "teacher" is an attacker who performs data poisoning attacks, and the learner is any intelligent system that adapts to inputs. In education the "student" is really a human student, and the teacher has a target model (i.e. the educational goal). If we are willing to assume a cognitive learning model of the student, we can use machine teaching to reverse-engineer the optimal training data -- which will be the optimal, personalized lesson for that student.

This page contains our research on the theory, algorithms, and applications of machine teaching.


Publications

Tutorials

Theory of Machine Teaching

Applications in Security, Trustworthy, and Interpretable AI

Applications in Human Computer Interaction

Applications in Cognitive Psychology and Education


Workshops


Talks


Code and Data


In the media

Back to Professor Zhu's home page