3 papers
cs.LG2023
Unbiased Learning of Deep Generative Models with Structured Discrete Representations
Harry Bendekgey, Gabriel Hope, Erik B. Sudderth
By composing graphical models with deep learning architectures, we learn generative models with the strengths of both frameworks. The structured variational autoencoder (SVAE) inhe…
cond-mat.stat-mech2020
Scaling study of diffusion in dynamic crowded spaces
H. Bendekgey, G. Huber, D. Yllanes
We formulate a scaling theory for the long-time diffusive motion in a space occluded by a high density of moving obstacles in dimensions 1, 2 and 3. Our tracers diffuse anomalously…
stat.AP2018
Consistency of Forecasts for the U.S. House of Representatives
Henry Bendekgey
We consider the performance of the foremost academic House of Representatives forecasting models in the 2018 elections. In creating open-source implementations of these models, we…