We introduce Dirichlet pruning, a novel post-processing technique to transform a large neural network model into a compressed one. Dirichlet pruning is a form of structured pruning which assigns the
Physics-constrained Deep Learning of Multi-zone Building Thermal Dynamics. (arXiv:2011.05987v1 [cs.LG])
We present a physics-constrained control-oriented deep learning method for modeling building thermal dynamics. The proposed method is based on the systematic encoding of physics-based prior knowledge into a structured recurrent
Maximum sampled conditional likelihood for informative subsampling. (arXiv:2011.05988v1 [math.ST])
Subsampling is a computationally effective approach to extract information from massive data sets when computing resources are limited. After a subsample is taken from the full data, most available methods
Linear Dilation-Erosion Perceptron for Binary Classification. (arXiv:2011.05989v1 [cs.LG])
In this work, we briefly revise the reduced dilation-erosion perceptron (r-DEP) models for binary classification tasks. Then, we present the so-called linear dilation-erosion perceptron (l-DEP), in which a linear transformation
Solving high-dimensional parameter inference: marginal posterior densities & Moment Networks. (arXiv:2011.05991v1 [stat.ML])
High-dimensional probability density estimation for inference suffers from the “curse of dimensionality”. For many physical inference problems, the full posterior distribution is unwieldy and seldom used in practice. Instead, we
Towards NNGP-guided Neural Architecture Search. (arXiv:2011.06006v1 [cs.LG])
The predictions of wide Bayesian neural networks are described by a Gaussian process, known as the Neural Network Gaussian Process (NNGP). Analytic forms for NNGP kernels are known for many
Understanding College Students’ Phone Call Behaviors Towards a Sustainable Mobile Health and Wellbeing Solution. (arXiv:2011.06007v1 [cs.HC])
During the transition from high school to on-campus college life, a student leaves home and starts facing enormous life changes, including meeting new people, more responsibilities, being away from family,
Football tracking networks: Beyond event-based connectivity. (arXiv:2011.06014v1 [cs.SI])
We propose using Network Science as a complementary tool to analyze player and team behavior during a football match. Specifically, we introduce four kinds of networks based on different ways
GANMEX: One-vs-One Attributions using GAN-based Model Explainability. (arXiv:2011.06015v1 [cs.LG])
Attribution methods have been shown as promising approaches for identifying key features that led to learned model predictions. While most existing attribution methods rely on a baseline input for performing
Policing Chronic and Temporary Hot Spots of Violent Crime: A Controlled Field Experiment. (arXiv:2011.06019v1 [stat.AP])
Hot-spot-based policing programs aim to deter crime through increased proactive patrols at high-crime locations. While most hot spot programs target easily identified chronic hot spots, we introduce models for predicting
