… small zones of cities (eg, single intersections) and cannot be used to gather insights for an
entire city. … In our work, we propose a reinforcement learning frameworks to overtake these two …

… We propose CityFlow, an efficient, multi-agent reinforcement learning environment for
large scale city traffic scenario. Researchers can use it as a testbed for traffic signal control …

… city. Recent advances primarily rely on deep reinforcement learning (DRL) to directly learn
… context-aware approach to improve the deep reinforcement learning based taxi dispatching. …

… To address these limitations, we propose a reinforcement learning based method for the city
metro … • We use real city-scale human mobility information to expand a metro network. The …

… In this paper, we propose a deep reinforcement learning method to tackle the problem of
city-level traffic signal control. We are the first to evaluate the RL-based traffic signal control …

… can be plausibly achieved using reinforcement learning and game-theoretic approaches [8].
Reinforcement learning (RL) has shown good potential for self-learning closed-loop optimal …

… Reinforcement learning (RL), formulated under the framework of Markov decision process
(MDP), is a promising alternative to learn ATSC based on real-world traffic measurements [8]. …

… We design ETA-Aware RL-Dispatch, a reinforcement learning algorithm with a carefully …
Based on the large-scale data, we build an emulator for training the RL model, and test the …

… ), we investigate how our model transfers between different scales of synthetic … -scale
community with a better final performance. It is worth noting that transferring to even larger scales is …

… a deep reinforcement learning (RL) approach to automatically learn prevention strategies
in … deep reinforcement learning is particularly interesting, as it allows us to set up a learning …