Deep Reinforcement Learning Network Research Articles

Feasibility Analysis and Application of Reinforcement Learning Algorithm Based on Dynamic Parameter Adjustment

Reinforcement learning, as a branch of machine learning, has been gradually applied in the control field. However, in the practical application of the algorithm, the hyperparametric approach to network settings for deep reinforcement learning still follows the empirical attempts of traditional machine learning (supervised learning and unsupervised learning). This method ignores part of the information generated by agents exploring the environment contained in the updating of the reinforcement learning value function, which will affect the performance of the convergence and cumulative return of reinforcement learning. The reinforcement learning algorithm based on dynamic parameter adjustment is a new method for setting learning rate parameters of deep reinforcement learning. Based on the traditional method of setting parameters for reinforcement learning, this method analyzes the advantages of different learning rates at different stages of reinforcement learning and dynamically adjusts the learning rates in combination with the temporal-difference (TD) error values to achieve the advantages of different learning rates in different stages to improve the rationality of the algorithm in practical application. At the same time, by combining the Robbins–Monro approximation algorithm and deep reinforcement learning algorithm, it is proved that the algorithm of dynamic regulation learning rate can theoretically meet the convergence requirements of the intelligent control algorithm. In the experiment, the effect of this method is analyzed through the continuous control scenario in the standard experimental environment of ”Car-on-The-Hill” of reinforcement learning, and it is verified that the new method can achieve better results than the traditional reinforcement learning in practical application. According to the model characteristics of the deep reinforcement learning, a more suitable setting method for the learning rate of the deep reinforcement learning network proposed. At the same time, the feasibility of the method has been proved both in theory and in the application. Therefore, the method of setting the learning rate parameter is worthy of further development and research.

A Survey on Deep Reinforcement Learning Network for Traffic Light Cycle Control

A Traffic signal control is a challenging problem and to minimize the travel time of vehicles by coordinating their movements at the road intersections. In recent years traffic signal control systems have on over simplified information and rule-based methods and we have large amounts of data, more computing power and advanced methods to drive the development of intelligent transportation. An intelligent transport system to use the machine learning methods likes reinforcement learning and to explain the acknowledged transportation approaches and a list of recent literature in traffic signal control. In this survey can foster interdisciplinary research on this important topic.

UAV navigation in high dynamic environments: A deep reinforcement learning approach

Unmanned Aerial Vehicle (UAV) navigation is aimed at guiding a UAV to the desired destinations along a collision-free and efficient path without human interventions, and it plays a crucial role in autonomous missions in harsh environments. The recently emerging Deep Reinforcement Learning (DRL) methods have shown promise for addressing the UAV navigation problem, but most of these methods cannot converge due to the massive amounts of interactive data when a UAV is navigating in high dynamic environments, where there are numerous obstacles moving fast. In this work, we propose an improved DRL-based method to tackle these fundamental limitations. To be specific, we develop a distributed DRL framework to decompose the UAV navigation task into two simpler sub-tasks, each of which is solved through the designed Long Short-Term Memory (LSTM) based DRL network by using only part of the interactive data. Furthermore, a clipped DRL loss function is proposed to closely stack the two sub-solutions into one integral for the UAV navigation problem. Extensive simulation results are provided to corroborate the superiority of the proposed method in terms of the convergence and effectiveness compared with those of the state-of-the-art DRL methods.

Goal-Oriented Obstacle Avoidance with Deep Reinforcement Learning in Continuous Action Space

In this paper, we propose a goal-oriented obstacle avoidance navigation system based on deep reinforcement learning that uses depth information in scenes, as well as goal position in polar coordinates as state inputs. The control signals for robot motion are output in a continuous action space. We devise a deep deterministic policy gradient network with the inclusion of depth-wise separable convolution layers to process the large amounts of sequential depth image information. The goal-oriented obstacle avoidance navigation is performed without prior knowledge of the environment or a map. We show that through the proposed deep reinforcement learning network, a goal-oriented collision avoidance model can be trained end-to-end without manual tuning or supervision by a human operator. We train our model in a simulation, and the resulting network is directly transferred to other environments. Experiments show the capability of the trained network to navigate safely around obstacles and arrive at the designated goal positions in the simulation, as well as in the real world. The proposed method exhibits higher reliability than the compared approaches when navigating around obstacles with complex shapes. The experiments show that the approach is capable of avoiding not only static, but also dynamic obstacles.

Collision avoidance for an unmanned surface vehicle using deep reinforcement learning

In this paper, a deep reinforcement learning (DRL)-based collision avoidance method is proposed for an unmanned surface vehicle (USV). This approach is applicable to the decision-making stage of collision avoidance, which determines whether the avoidance is necessary, and if so, determines the direction of the avoidance maneuver. To utilize the visual recognition capability of deep neural networks as a tool for analyzing the complex and ambiguous situations that are typically encountered, a grid map representation of the ship encounter situation was suggested. For the composition of the DRL network, we proposed a neural network architecture and semi-Markov decision process model that was specially designed for the USV collision avoidance problem. The proposed DRL network was trained through repeated simulations of collision avoidance. After the training process, the DRL network was implemented in collision avoidance experiments and simulations to evaluate its situation recognition and collision avoidance capability.

A Robust Context-Aware Proposal Refinement Method for Weakly Supervised Object Detection

Supervised object detection models require fully annotated data for training the network. However, labeling large datasets is a very time-consuming task, therefore, weakly supervised object detection (WSOD) is a substitute approach to fully supervised learning for the object detection task. Many methods have been proposed for WSOD to date, their performance is still lower than supervised approaches since WSOD is a very challenging task. The major problem with existing WSOD methods is partial object detection and false detection in an objects cluster with the same category. The majority of the methods on WSOD follow multiple instance learning approaches, which does not guarantee the completeness of detected objects. To address these issues, we propose a three-fold refinement strategy to proposals to learn complete instances. We generate class-specific localization maps by fused class activation maps obtained from fused complementary classification networks. These localization maps are used to amend the detected proposals from the instance classification branch (detection network). Deep reinforcement learning networks are proposed to learn decisive-agent and rectifying-agent based on policy gradient algorithm to further refine the proposals. The refined bounding boxes are then fed to instance classification network. The refinement operations result in learning complete objects and greatly improve detection performance. Experimental results show better detection performance by the proposed WSOD method compared to the state-of-the-art methods on PASCAL VOC2007 and VOC2012 benchmarks.

Reinforcement Learning With Low-Complexity Liquid State Machines.

We propose reinforcement learning on simple networks consisting of random connections of spiking neurons (both recurrent and feed-forward) that can learn complex tasks with very little trainable parameters. Such sparse and randomly interconnected recurrent spiking networks exhibit highly non-linear dynamics that transform the inputs into rich high-dimensional representations based on the current and past context. The random input representations can be efficiently interpreted by an output (or readout) layer with trainable parameters. Systematic initialization of the random connections and training of the readout layer using Q-learning algorithm enable such small random spiking networks to learn optimally and achieve the same learning efficiency as humans on complex reinforcement learning (RL) tasks like Atari games. In fact, the sparse recurrent connections cause these networks to retain fading memory of past inputs, thereby enabling them to perform temporal integration across successive RL time-steps and learn with partial state inputs. The spike-based approach using small random recurrent networks provides a computationally efficient alternative to state-of-the-art deep reinforcement learning networks with several layers of trainable parameters.

A Deep Reinforcement Learning Network for Traffic Light Cycle Control

Existing inefficient traffic light control causes numerous problems, such as long delay and waste of energy. To improve efficiency, taking real-time traffic information as an input and dynamically adjusting the traffic light duration accordingly is a must. In terms of how to dynamically adjust traffic signals' duration, existing works either split the traffic signal into equal duration or extract limited traffic information from the real data. In this paper, we study how to decide the traffic signals' duration based on the collected data from different sensors and vehicular networks. We propose a deep reinforcement learning model to control the traffic light. In the model, we quantify the complex traffic scenario as states by collecting data and dividing the whole intersection into small grids. The timing changes of a traffic light are the actions, which are modeled as a high-dimension Markov decision process. The reward is the cumulative waiting time difference between two cycles. To solve the model, a convolutional neural network is employed to map the states to rewards. The proposed model is composed of several components to improve the performance, such as dueling network, target network, double Q-learning network, and prioritized experience replay. We evaluate our model via simulation in the Simulation of Urban MObility (SUMO) in a vehicular network, and the simulation results show the efficiency of our model in controlling traffic lights.

Multi-agent deep learning for simultaneous optimization for time and energy in distributed routing system

Packet routing problem most commonly emerges in the context of computer networks, thus the majority of routing algorithms existing nowadays is designed specifically for routing in computer networks. However, in the logistics domain, many problems can be formulated in terms of packet routing, e.g. in automated traffic routing or material handling systems. In this paper, we propose an algorithm for packet routing in such heterogeneous environments. Our approach is based on deep reinforcement learning networks combined with link-state protocol and preliminary supervised learning. Similarly to most routing algorithms, the proposed algorithm is a distributed one and is designed to run on a network constructed from interconnected routers. Unlike most other algorithms, proposed one views routers as learning agents, representing the routing problem as a multi-agent reinforcement learning problem. Modeling each router as a deep neural network allows each router to account for heterogeneous data about its environment, allowing for optimization of more complex cost functions, like in case of simultaneous optimization of bag delivery time and energy consumption in a baggage handling system. We tested the algorithm using manually constructed simulation models of computer network and baggage handling system. It outperforms state-of-the-art routing algorithms.

A Deep Hierarchical Approach to Lifelong Learning in Minecraft

We propose a lifelong learning system that has the ability to reuse and transfer knowledge from one task to another while efficiently retaining the previously learned knowledge-base. Knowledge is transferred by learning reusable skills to solve tasks in Minecraft, a popular video game which is an unsolved and high-dimensional lifelong learning problem. These reusable skills, which we refer to as Deep Skill Networks, are then incorporated into our novel Hierarchical Deep Reinforcement Learning Network (H-DRLN) architecture using two techniques: (1) a deep skill array and (2) skill distillation, our novel variation of policy distillation (Rusu et. al. 2015) for learning skills. Skill distillation enables the H-DRLN to efficiently retain knowledge and therefore scale in lifelong learning, by accumulating knowledge and encapsulating multiple reusable skills into a single distilled network. The H-DRLN exhibits superior performance and lower learning sample complexity compared to the regular Deep Q Network (Mnih et. al. 2015) in sub-domains of Minecraft.