Unsupervised machine-learning techniques for high-energy collider searches

Machine-learning techniques are quickly becoming important tools in physics analyses at the LHC, with both the experimental collaborations and theorists investing much time and effort into the design of these tools.  In this talk I will present results from my own work on unsupervised machine-learning techniques (1904.04200 and 2005.12319), along with an overview of other important work in the field.  I will discuss the successes and the challenges of this line of research in general, and how the latter could be addressed in future.