This paper considers bidding behavior in a repeated procurement auction setting. We study highway procurement data for the state of California between December 1994 and October 1998. We consider a dynamic bidding model that takes into account the presence of intertemporal constraints such as capacity constraints. We estimate the model non-parametrically and assess the presence of dynamic constr...

We address issues related to the use of simulation models to assess highway capacity and quality of service. This topic is of great interest currently given the growing trend toward the use of simulation for such purposes. Descriptive paradigms, continuity, clarity, reliability, repeatability, and consistency with theory are among the issues examined. Four paradigms for describing simulation mo...

While machine learning models are powering more and everyday devices, there is a growing need for explaining them. This especially applies to the use of deep reinforcement in solutions that require security, such as vehicle motion planning. In this paper, we propose method understanding what RL agent’s decision based on. The relies on conducting statistical analysis massive set state-decisions ...

We study the traffic characteristics on a single-lane highway with a slowdown section using the deterministic cellular automaton (CA) model. Based on the theoretical analysis, the relationships among local mean densities, velocities, traffic fluxes, and global densities are derived. The results show that two critical densities exist in the evolutionary process of traffic state, and they are sig...

Recently, the rapid development of word embedding and neural networks has brought new inspiration to various NLP and IR tasks. In this paper, we describe a staged hybrid model combining Recurrent Convolutional Neural Networks (RCNN) with highway layers. The highway network module is incorporated in the middle takes the output of the bidirectional Recurrent Neural Network (Bi-RNN) module in the ...