Machine learning techniques benefit science and industry primarily insofar as they enable data analysts to better understand their data, make valid conclusions, and gain insight into their domain of investigation. Studies in anthropocentric data analysis seek to understand how human judgment is applied in the data analysis process and to develop methods that provide explicit opportunities for human interaction and insight. We present three studies of human visual reasoning exploring the extent to which novice and expert subjects are able to judge the quality of and understand cluster analysis and dimensionality reduction stimuli. We then investigate whether humans are able to learn how cluster analysis algorithms function simply through interacting with them rapidly across diverse data sets. Finally we show how multicore processing on CPUs and GPUs can move human/algorithm interactions closer to real-time

Abstract—Gaze following, the ability to redirect one’s visual at- tention to look at what another person is seeing, is foundational for imitation, word learning, and theory-of-mind. Previous theories have suggested that the development of gaze following in human infants is the product of a basic gaze following mechanism, plus the gradual incorporation of several distinct new mechanisms that im- prove the skill, such as spatial inference, and the ability to use eye direction information as well as head direction. In this paper, we offer an alternative explanation based on a single learning mecha- nism. From a starting state with no knowledge of the implications of another organism’s gaze direction, our model learns to follow gaze by being placed in a simulated environment where an adult caregiver looks around at objects. Our infant model matches the development of gaze following in human infants as measured in key experiments that we replicate and analyze in detail.

In many problem domains data may come from multiple sources (or views), such as video and audio from a camera or text on and links to a web page. These multiple views of the data are often not directly comparable to one another, and thus a principled method for their integration is warranted. In this paper we develop a new algorithm to leverage information from multiple views for unsupervised clustering by constructing a custom kernel. We generate a multipartite graph (with the number of parts given by the number of views) that induces a kernel we then use for spectral clustering. Our algorithm can be seen as a generalization of co-clustering and spectral clustering and a relative of Kernel Canonical Correlation Analysis. We demonstrate the algorithm on four data sets: an illustrative artificial data set, synthetic fMRI data, voxels from an fMRI study, and a collection of web pages. Finally, we compare its performance to common alternatives.