Tomas Singliar

Learning to detect incidents from noisily labeled data

Many deployed traffic incident detection systems use algorithms that require significant manual tuning. We seek machine learning incident detection solutions that reduce the need for manual adjustments by taking advantage of massive databases of traffic sensor measurements. We first examine which traffic flow features are most useful for the incident detection task. Then we show that a supervised learner based on the SVM model outperforms a fixed detection model used by state-of-the-art traffic incident detectors. However, the performance of a supervised learner suffers from temporal noise in the data labels due to imperfections of the incident logging procedure. Correcting these misaligned incident times in the training data achieves further improvements in detection performance. We propose a label realignment model based on a dynamic Bayesian network to re-estimate the correct position (time) of the incident in the data. Training on the automatically realigned data then consistently leads to improved detection performance in the low false positive region.

Modeling Highway Traffic Volumes

Most traffic management and optimization tasks, such as accident detection or optimal vehicle routing, require an ability to adequately model, reason about and predict irregular and stochastic behavior. Our goal is to create a probabilistic model of traffic flows on highway networks that is realistic from the point of applications and at the same time supports efficient learning and inference. We study several multivariate probabilistic models and analyze their respective strengths. To balance accuracy and efficiency, we propose a novel learning model, mixture of Gaussian trees, and show its advantages in learning and inference. All models are evaluated on real-world traffic flow data from highways of the Pittsburgh area.

On synchronicity and concurrency in Petri nets

In the paper we extend the algebraic description of Petri nets based on rewriting logic by introducing a partial synchronous operation in order to distinguish between synchronous and concurrent occurrences of transitions. In such an extension one first needs to generate steps of transitions using a partial operation of synchronous composition and then to use these steps to generate process terms using partial operations of concurrent and sequential composition. Further, we define which steps are true synchronous. In terms of causal relationships, such an extension corresponds to the approach described in [6,7,9], where two kinds of causalities are defined, first saying (as usual) which transitions cannot occur earlier than others, while the second indicating which transitions cannot occur later than others. We illustrate this claim by proving a one-to-one correspondence between such extended algebraic semantics of elementary nets with inhibitor arcs and causal semantics of elementary nets with inhibitor arcs presented in [7].

Learning to Detect Adverse Traffic Events from Noisily Labeled Data

Many deployed traffic incident detection systems use algorithms that require significant manual tuning. We seek machine learning incident detection solutions that reduce the need for manual adjustments by taking advantage of massive databases of traffic sensor network measurements. First, we show that a rather straightforward supervised learner based on the SVM model outperforms a fixed detection model used by state-of-the-art traffic incident detectors. Second, we seek further improvements of learning performance by correcting misaligned incident times in the training data. The misalignment is due to an imperfect incident logging procedure. We propose a label realignment model based on a dynamic Bayesian network to re-estimate the correct position (time) of the incident in the data. Training on the automatically realigned data consistently leads to improved detection performance in the low false positive region.