Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for reinforcement learning using a partitioned reinforcement learning input state space (RL input state space). One of the methods includes maintaining data defining a plurality of partitions of a space of reinforcement learning (RL) input states, each partition corresponding to a respective supervised learning model; obtaining a current state representation that represents a current state of the environment; for the current state representation and for each action in the set of actions, identifying a respective partition and processing the action and the current state representation using the supervised learning model that corresponds to the respective partition to generate a respective current value function estimate; and selecting an action to be performed by the computer-implemented agent in response to the current state representation using the respective current value function estimates.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for reinforcement learning using confidence scores. One of the methods includes receiving a current observation; for each of multiple actions: determining a respective value function estimate that is an estimate of a return resulting from the agent performing the action in response to the current observation, determining a respective confidence score that is a measure of confidence that the respective value function estimate for the action is an accurate estimate of the return that will result from the agent performing the action in response to the current observation, adjusting the respective value function estimate for the action using the respective confidence score for the action to determine a respective adjusted value function estimate; and selecting an action to be performed by the agent in response to the current observation using the respective adjusted value function estimates.

Abstract: A traffic simulator for simulating traffic events in a network, in which the events behave according to one or more statistical models. One or more event sources, are used for randomly issuing one or more events at discrete time slots within a predefined maximum event period. If only one of the sources issues an event at specific time slot, the event is output into an Event Labeler, and if more than one event is being issued at a specific time slot, one of the events is selected and output into the Event Labeler and the rest of the events are postponed to the next time slot. In case more than one event is postponed to a next time slot, one of the postponed event, or possibly a newly issued event, is selected in the next time slot according to a predetermined output selection policy. The selected event is then output to the Event Labeler. An Event Labeler is used to label a destination, being randomly selected from a given list of destinations, to each output event.