Zhi, Weifeng (University of California, Davis) | Wang, Xiang (University of California, Davis) | Qian, Buyue (University of California, Davis) | Butler, Patrick (Virginia Tech) | Ramakrishnan, Naren (Virginia Tech) | Davidson, Ian (University of California, Davis)
Clustering with constraints is an important and developing area. However, most work is confined to conjunctions of simple together and apart constraints which limit their usability. In this paper, we propose a new formulation of constrained clustering that is able to incorporate not only existing types of constraints but also more complex logical combinations beyond conjunctions. We first show how any statement in conjunctive normal form (CNF) can be represented as a linear inequality. Since existing clustering formulations such as spectral clustering cannot easily incorporate these linear inequalities, we propose a quadratic programming (QP) clustering formulation to accommodate them. This new formulation allows us to have much more complex guidance in clustering. We demonstrate the effectiveness of our approach in two applications on text and personal information management. We also compare our algorithm against existing constrained spectral clustering algorithm to show its efficiency in computational time.
The news: The UK Home Office has said it will stop using an algorithm to process visa applications that critics claim is racially biased. Opponents to it argue that the algorithm's use of nationality to decide which applications get fast-tracked has led to a system in which "people from rich white countries get "Speedy Boarding"; poorer people of color get pushed to the back of the queue." Time for a redesign: The Home Office denies that its system is racially biased and litigation is still ongoing. Even so, the Home Office has agreed to drop the algorithm and plans to relaunch a redesigned version later this year, after conducting a full review that will look for unconscious bias. In the meantime the UK will adopt a temporary system that does not use nationality to sort applications.
We study the problem of decision-theoretic online learning (DTOL). Motivated by practical applications, we focus on DTOL when the number of actions is very large. Previous algorithms for learning in this framework have a tunable learning rate parameter, and a major barrier to using online-learning in practical applications is that it is not understood how to set this parameter optimally, particularly when the number of actions is large. In this paper, we offer a clean solution by proposing a novel and completely parameter-free algorithm for DTOL. In addition, we introduce a new notion of regret, which is more natural for applications with a large number of actions.
Artificial Intelligence is considered a top-notch technology that has the potential to solve many of humanity's problems. Be it healthcare, education, or finance, it is looked upon as a benediction, as it carries promise for solutions, which evaded humans for centuries. The human mind is curious and wants to experiment to test the limits of technology. In this endeavour, we have some unusual yet innovative applications of AI, which might evoke some curiosity even for a tech-averse individual. Artificial intelligence has started judging a beauty contest!