Natural Actor-Critic for Road Traffic Optimisation

By Silvia Richter, Douglas Aberdeen, Jin Yu. 2006

112 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Automotive Engineering, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Automotive Engineering

In Control and Systems Engineering

Citations in the corpus

Multiagent Learning: Basics, Challenges, and Prospects (2012, background): “…the integration of learning in these real-world domains and has shown promising results that justify a stronger focus by the community on complex systems; for example, see Richter, Aberdeen, and Yu (2006); Kalyanakrish-nan, Liu, and Stone (2006); Abbeel et al. (2007); Agogino and Tumer (2012).”
Multiagent Reinforcement Learning for Integrated Network of Adaptive Traffic Signal Controllers (MARLIN-ATSC): Methodology and Large-Scale Application on Downtown Toronto (2013, background): “However, the applicability of such approaches is limited to optimize a few traffic signal agents due to the obvious exponentially increasing joint space of states with the increase in the number of agents [8].”
Ieee Transactions on Systems, Man and Cybernetics—part C: Applications and Reviews 1 a Survey of Actor-critic Reinforcement Learning: Standard and Natural Policy Gradients (unknown): “Barto et al. [46], FACRLN [53], [54], CACM [55], Bhatnagar [56], Chun-Gui et al. [57], Kimura et al. [58], Raju et al. [59] (e)NAC [16], [52], Park et al. [60], Girgin and Preux [61], Kimura [62],Richter et al. [2], Kim et al. [63], Nakamura et al. [64], El-Fakdi et al. [65] Average return”
Variational Bayesian learning of nonlinear hidden state-space models for model predictive control (2009, background): “Learning is extremely important for control of complex systems [2] ranging from anthropomorphic robots [23] over military aircrafts in hazardous environments [10] to road traffic optimisation [28].”
Online Model Learning Algorithms for Actor-Critic Control (2015, methodology): “The benefit of a parameterised policy is that a spectrum of continuous actions can be generated, but the optimisation methods used (typically called policy gradient methods) suffer from high variance in the estimates of the gradient, leading to slow learning (Baxter and Bartlett, 2001; Berenji and V”
Natural Actor-Critic (2003, background): “The current range of additional applications includes optimization of constrained reaching movements of humanoid robots [12], traffic-light system optimization [23], multi-agent system optimization [11,28], conditional random fields [27] and gait optimization in robot locomotion [26,17].”
Reinforcement learning of motor skills with policy gradients (2008, background): “This class of algorithms has been widely accepted by now and has been applied in a variety of settings (Guenter, Hersch, Calinon, & Billard, in press; Mori et al., 2004; Mori, Nakamura, & Ishii, 2005; Nakamura et al., 2004; Park, Kim, & Kang, 2005; Richter, Aberdeen, & Yu, 2007; Sato et al., 2002;”