Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

The Rise of the Creative Class

Inclusion and democracy.

Research in graph signal processing (GSP) aims to develop tools for processing data defined on irregular graph domains. In this paper, we first provide an overview of core ideas in GSP and their connection to conventional digital signal processing, along with a brief historical perspective to highlight how concepts recently developed in GSP build on top of prior research in other areas. We then summarize recent advances in developing basic GSP tools, including methods for sampling, filtering, or graph learning. Next, we review progress in several application areas using GSP, including processing and analysis of sensor network data, biological data, and applications to image processing and machine learning.

Graph Signal Processing: Overview, Challenges, and Applications

Sampson, Robert J. 2012. Great American city: Chicago and the enduring neighborhood effect. Chicago: University of Chicago Press. ISBN-13: 9780226734569. pp. 552, $27.50 cloth. Robert J. Sampson’s ...

Great American city: Chicago and the enduring neighborhood effect

The Spatial Diffusion of Green Building Policies

Of the three primary components of housing affordability measures - housing cost, transportation costs, and utility costs - utility expenditures are the least understood, yet the one area where the cost burden can be reduced without household relocation. Existing data sources to measure energy costs are limited to surveys with small samples and low spatial and temporal resolution, such as the American Housing Survey and the Residential Energy Consumption Survey. This paper presents a new method to estimate household energy cost burdens that leverages actual building energy use data for 13,000 multi-family properties across six U.S. cities, and links energy costs to savings opportunities based on an analysis of 3,000 energy audit reports. We examine differentials in cost burdens across household demographic and socioeconomic characteristics, and analyze spatial, regional, and building-level variations in energy use and expenditures. We find that the average low-income household has an energy cost burden of 7%, whereas higher-income households have an average burden of 2%. Notably, we find that even within defined income bands, minority households experience higher energy cost burdens than non-Hispanic White households. For lower-income households, low-cost energy improvements could reduce energy costs by as much as $1,500 per year.

Energy Cost Burdens for Low-Income and Minority Households in Six U.S. Cities: Evidence from Energy Benchmarking and Audit Data

The widespread adoption and use of mobile phone and Wi-Fi enabled devices has created new sources of big data useful for understanding urban mobility and real-time population activity. Wi-Fi networks in particular are being increasingly deployed throughout dense urban areas providing new possibilities to study the dynamics of urban populations. However, there is little known about the ability of Wi-Fi networks to quantify urban populations in real-time, including the accuracy, coverage and capture rate of these networks. In this research, we compare three methods for generating population estimates using Wi-Fi probe request data and validate these approaches with manual count data for a two month period. Despite concerns related to data bias and representativeness, we find that Wi-Fi data can be a viable source of information for estimating urban populations.

Validating the Use of Wi-Fi Signals to Estimate Hyperlocal Urban Populations

New York City’s Energy and Water Use Report

Mandatory building energy audits in New York City lead to a modest reduction of energy use of 2.5% and 4.9% for multifamily and office buildings, respectively. This suggests that other approaches besides mandatory audit policies, such as building energy grading and carbon intensity targets, may be needed to achieve sustainability goals.

Constantine E. Kontokosta

Papers

The Spatial Diffusion of Green Building Policies

Energy Cost Burdens for Low-Income and Minority Households in Six U.S. Cities: Evidence from Energy Benchmarking and Audit Data

Validating the Use of Wi-Fi Signals to Estimate Hyperlocal Urban Populations

New York City’s Energy and Water Use Report

Mandatory building energy audits alone are insufficient to meet climate goals