Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 184 for “"DQN"”.
-
A vision based DQN exploration algorithm for collision avoidance
… introduce a new variation of the Deep Q-Network (DQN) algorithm for UAV collision avoidance. Exploration with other variations of DQN for collision avoidance such as D3QN, are typically done through uniform sampling of actions, however, the challenge is environments inherently have sparse rewards …
-
DQN-Guided Rule-Class Selection for Search-Based Query Rewriting
… decision problem and proposes DQRS (DQN-Guided Rule-Class Selection), which ranks the legal rule classes at each state with a Deep Q-Network and forwards only the top k to the unchanged rewriter routine. On an in-domain workload derived from the TPC-H benchmark at k = 1, average …
-
Kernel-Controlled DQN Based CNN Pruning for Model Compression and Acceleration
Apart from the accuracy, the size of convolutional neural networks (CNN) models is another principal factor for facilitating the deployment of models on memory, power and budget constrained devices. However, conventional model compression techniques require human experts to setup parameters to …
-
Realtime Vehicle Route Optimisation via DQN for Sustainable and Resilient Urban Transportation Network
Traffic congestion has become one of the most serious contemporary city issues for urban transportation network as it leads to unnecessary high energy consumption, air pollution and extra travelling time. During this decade, many optimization algorithms have been designed to achieve the optimal …
-
A novel Target-DQN model for handover control parameters optimization in ultra-dense cellular networks
… This thesis presents a Deep Q-Network (DQN)-based framework for optimizing handover control parameters (HCPs), including Time-to-Trigger (TTT), Hysteresis (Hys), and A3Offset. The proposed Target-DQN (TDQN) model incorporates a tunable Search Radius (SR) in its action definition to …
-
AI-řízené semafory Brna
… metody: pevné časování, max-pressure, nezávislé DQN a koordinované GAT-DQN inspirované architekturou CoLight. Experimenty ukazují, že DQN i GAT-DQN snižují průměrné čekací doby přibližně o 60 % oproti pevnému časování, přičemž DQN dosahuje nejlepších výsledků na čtyřech z pěti dopravních profilů …
-
Quantification learning with deep neural networks
… end-to-end Deep Quantification Neural Network (DQN) framework. DQN jointly learns effective feature representations and class distribution estimates. We formulate the quantification learning problem as a maximum likelihood problem. DQN can be seen as learning a mapping function of the input, a …
-
AI-řízené semafory Brna
… a dva řadiče založené na posilovaném učení (DQN a GAT-DQN), vyhodnocené na ručně nakonfigurované podsíti centra Brna se syntetickými dopravními profily. Všechny tři adaptivní řídicí strategie výrazně snižují simulovanou čekací dobu oproti pevnému časování, avšak při shodném rozhodovacím …
-
Novel reaction-based De Novo molecular design in 2D & 3D using deep Q-Learning
… in particular. Molecule Deep Q-Networks (MolDQN) is a de novo algorithm developed by Zhou et al. in 2019 that combines simple chemical actions – adding or removing atoms or bonds – together with reinforcement learning. Consequently, MolDQN enables “local” search of chemical space by enabling …
-
Mastering Cooperative, Incomplete Information Board Games by Self-Play
… to apply the techniques of Deep Q-Networks (DQN) and Deep Monte Carlo (DMC) to a board-card games case study. The agents learn from self-play only. RLCard was chosen as the RL research framework.<br/>The results with DMC are much better than with DQN. DQN scores a maximum win rate of 68% …
-
Multi-Agent Deep Reinforcement Learning Assisted Pre-connect Handover Management
… reliable handover for 5G networks. PHO utilizes DQN algorithm to facilitate the sequential decision-making problem of target base station (BS) selection based on the RSRQs and RSRQ change rates of all the surrounding candidate BSs. A MADRL solution is tailored to extend the DQN-assisted …
-
Deep Q-Learning pritaikymas judančio veikėjo sekimui 2D gaudynių žaidime naudojant vidinį LiDAR žemėlapį /
In this thesis the Rainbow DeepQ Network (DQN) algorithm is tested in a custommade 2D platformer tag game with three different state representations: Full Screenshot, LiDAR map with persistent player state knowledge (oracle) and LiDAR map with a line of sight restriction. The algo rithm is …
-
Comparative study on classical and modern ways of traffic signal control with the use of a simulator
… o Max Pressure και ένα μοντέλο μηχανικής μάθησης DQN - σε διαφορετικά σενάρια κυκλοφορίας. Η υλοποίηση των παραδοσιακών μεθόδων είναι σχετικά απλή. Η μέθοδος Webster έχει ως στόχο στην ελαχιστοποίηση της καθυστέρησης των οχημάτων προσαρμόζοντας το μήκος του κύκλου των σηματοδοτών κυκλοφορίας βάσει …
-
Adaptive traffic signal control using deep reinforcement learning for network traffic incidents
… with a single intersection scenario, where the DQN algorithm was modeled to form the RL agent traffic signal controller. To improve the training process and model performance, experience replay and target network were implemented to solve the limitations of DQN. Hyperparameter tuning was …
-
OPTIMAL NAVAL MOVEMENT SIMULATION WITH REINFORCEMENT LEARNING AI AGENTS
… in six scenarios. Evaluations of Deep Q-Network (DQN), Monte Carlo tree search (MCTS), and AlphaStar AI agents across various scenarios revealed that DQN and MCTS agents were able to identify superior strategies, with DQN demonstrating consistently high scores and outperforming human players in …
-
Optimal Traffic Flow Control Strategies, on a Lane Group- and Vehicle-Based Level, at Freeway Lane-Drops under the Environment of Connected and Automated Vehicles
… control strategy is based on the Deep Q-Network (DQN) Reinforcement Learning (RL) algorithm, and is designed to determine the driving speed and lane-change maneuvers that would enable the CAV to go through the bottleneck, with the least amount of delay. The DQN RL agent was trained using the …
-
Visual Navigation of Wheeled Mobile Robots Using Deep Reinforcement Learning
… namely, value-learning based deep Q-network (DQN) and policy gradient based asynchronous advantage actor critic (A3C) have been considered in this study. Both DRL algorithms have been implemented using RGB and depth images as inputs to generate outputs for the WMR for autonomous navigation in …
-
Deep Reinforcement Learning For The Computation Offloading In MIMO-based Edge Computing
… environment dynamics. First, a Deep Q-Network (DQN)-based algorithm to solve the curse of the state space explosion is defined. Then, a more general Proximal Policy Optimization (PPO)-based algorithm to solve the problem of discrete action space is introduced. Simulation results show that our …
-
Multi-agent deep reinforcement learning for Robocup Rescue Simulator
… how a DRL algorithm, termed Deep Q-Networks (DQN), can be applied to a collection of Atari 2600 games to surpass the performance of all previous algorithms and achieve a level that is comparable to a professional player. Their trained model received only raw pixels and game score as inputs to …
-
Creation of a Cognitive Radar with Machine Learning: Simulation and Implementation
… radar. We demonstrate that the Deep-Q Network (DQN) we construct is able to successfully avoid interfering with the communication system to improve its performance. We also show that the DQN method outperforms previous methods in terms of memory and handling new situations. In this thesis we …
Page 1 of 10