Rajesh Ranganath

Learn More
The neural patterns recorded during a neuroscientific experiment reflect complex interactions between many brain regions, each comprising millions of neurons. However, the measurements themselves are typically abstracted from that underlying structure. For example, functional magnetic resonance imaging (fMRI) datasets comprise a time series of(More)
BACKGROUND As adoption of electronic health records continues to increase, there is an opportunity to incorporate clinical documentation as well as laboratory values and demographics into risk prediction modeling. OBJECTIVE The authors develop a risk prediction model for chronic kidney disease (CKD) progression from stage III to stage IV that includes(More)
Appendix Gamma Distribution Figure 1 visually demonstrates how the sparse gamma distrubution and Pois-son distribution change when moving from low to high mean. We plot both the Poisson and sparse gamma distribution in both settings. The mode of the Pois-son distribution moves, while the slab grows for the gamma. General Algorithm. Following the notation(More)
Sarcomatoid carcinomas are biphasic tumors, which can occur at any site in the human body. Very few cases have been reported in the literature as arising from the penis. A few studies consider these tumors as a variant of squamous cell carcinoma or a metaplastic differentiation of the mesenchyme. Their clinical behavior is aggressive with both blood borne(More)
Stochastic variational inference (SVI) enables approximate posterior inference with large data sets for otherwise intractable models, but like all vari-ational inference algorithms it suffers from local optima. Deterministic annealing, which we formulate here for the generic class of conditionally conjugate exponential family models, uses a temperature(More)
  • 1