We develop a deep learning model to predict traffic flows. The main contribution is development of an architecture that combines a linear model that is fitted using 1 regularization and a sequence of tanh layers. The challenge of predicting traffic flows are the sharp nonlinearities due to transitions between free flow, breakdown, recovery and congestion. We show that deep learning architectures can capture these nonlinear spatio-temporal effects. The first layer identifies spatio-temporal relations among predictors and other layers model nonlinear relations. We illustrate our methodology on road sensor data from Interstate I-55 and predict traffic flows during two special events; a Chicago Bears football game and an extreme snowstorm event. Both cases have sharp traffic flow regime changes, occurring very suddenly, and we show how deep learning provides precise short term traffic flow predictions.

No abstract

Deep learning is a form of machine learning for nonlinear high dimensional pattern matching and prediction. By taking a Bayesian probabilistic perspective, we provide a number of insights into more efficient algorithms for optimisation and hyper-parameter tuning. Traditional high-dimensional data reduction techniques, such as principal component analysis (PCA), partial least squares (PLS), reduced rank regression (RRR), projection pursuit regression (PPR) are all shown to be shallow learners. Their deep learning counterparts exploit multiple deep layers of data reduction which provide predictive performance gains. Stochastic gradient descent (SGD) training optimisation and Dropout (DO) regularization provide estimation and variable selection. Bayesian regularization is central to finding weights and connections in networks to optimize the predictive bias-variance trade-off. To illustrate our methodology, we provide an analysis of international bookings on Airbnb. Finally, we conclude with directions for future research.

Deep learning applies hierarchical layers of hidden variables to construct nonlinear high dimensional predictors. Our goal is to develop and train deep learning architectures for spatio-temporal modeling. Training a deep architecture is achieved by stochastic gradient descent and dropout for parameter regularization with a goal of minimizing out-of-sample predictive mean squared error. To illustrate our methodology, we first predict the sharp discontinuities in traffic flow data, and secondly, we develop a classification rule to predict short-term futures market prices using order book depth. Finally, we conclude with directions for future research.

Connected–automated vehicle (CAV) technologies are likely to have significant effects not only on how vehicles operate in the transportation system, but also on how individuals behave and use their vehicles. While many CAV technologies—such as connected adaptive cruise control and ecosignals—have the potential to increase network throughput and efficiency, many of these same technologies have a secondary effect of reducing driver burden, which can drive changes in travel behavior. Such changes in travel behavior—in effect, lowering the cost of driving—have the potential to increase greatly the utilization of the transportation system with concurrent negative externalities, such as congestion, energy use, and emissions, working against the positive effects on the transportation system resulting from increased capacity. To date, few studies have analyzed the potential effects on CAV technologies from a systems perspective; studies often focus on gains and losses to an individual vehicle, at a single intersection, or along a corridor. However, travel demand and traffic flow constitute a complex, adaptive, nonlinear system. Therefore, in this study, an advanced transportation systems simulation model—POLARIS—was used. POLARIS includes cosimulation of travel behavior and traffic flow to study the potential effects of several CAV technologies at the regional level. Various technology penetration levels and changes in travel time sensitivity have been analyzed to determine a potential range of effects on vehicle miles traveled from various CAV technologies.

Platooning vehicles-connected and automated vehicles traveling with small intervehicle distances-use less fuel because of reduced aerodynamic drag. Given a network defined by vertex and edge sets and a set of vehicles with origin/destination nodes/times, we model and solve the combinatorial optimization problem of coordinated routing of vehicles in a manner that routes them to their destination on time while using the least amount of fuel. Common approaches decompose the platoon coordination and vehicle routing into separate problems. Our model addresses both problems simultaneously to obtain the best solution. We use modern modeling techniques and constraints implied from analyzing the platoon routing problem to address larger numbers of vehicles and larger networks than previously considered. While the numerical method used is unable to certify optimality for candidate solutions to all networks and parameters considered, we obtain excellent solutions in approximately one minute for much larger networks and vehicle sets than previously considered in the literature.

In many cases, an organization wishes to release some data, but is restricted in the amount of data to be released due to legal, privacy and other concerns. For instance, the US Census Bureau releases only 1% of its table of records every year, along with statistics about the entire table. However, the machine learning (ML) models trained on the released sub-table are usually sub-optimal. In this paper, our goal is to find a way to augment the sub-table by generating a synthetic table from the released sub-table, under the constraints that the generated synthetic table (i) has similar statistics as the entire table, and (ii) preserves the functional dependencies of the released sub-table. We propose a novel generative adversarial network framework called ITS-GAN, where both the generator and the discriminator are specifically designed to satisfy these two constraints. By evaluating the augmentation performance of ITS-GAN on two representative datasets, the US Census Bureau data and US Bureau of Transportation Statistics (BTS) data, we show that ITS-GAN yields high quality classification results, and significantly outperforms various state-of-the-art data augmentation approaches.

Understanding driving behaviors is essential for improving safety and mobility of our transportation systems. Data is usually collected via simulator-based studies or naturalistic driving studies. Those techniques allow for understanding relations between demographics, road conditions and safety. On the other hand, they are very costly and time consuming. Thanks to the ubiquity of smartphones, we have an opportunity to substantially complement more traditional data collection techniques with data extracted from phone sensors, such as GPS, accelerometer gyroscope and camera. We developed statistical models that provided insight into driver behavior in the San Francisco metro area based on tens of thousands of driver logs. We used novel data sources to support our work. We used cell phone sensor data drawn from five hundred drivers in San Francisco to understand the speed of traffic across the city as well as the maneuvers of drivers in different areas. Specifically, we clustered drivers based on their driving behavior. We looked at driver norms by street and flagged driving behaviors that deviated from the norm.