لطفا منتظر بمانید ...
0% Complete
صفحه اصلی
/
سی و سومین کنفرانس بین المللی مهندسی برق
Better Exploration In Single-Agent Q-Learning Using Controlled Linear Perturbation
نویسندگان :
Sadredin Hokmi
1
Mohammad Haeri
2
1- Sharif university of technology
2- Sharif university of technology
کلمات کلیدی :
Q-learning،Exploration،Controlled Linear perturbation،Convergence rate،Maze،Cart-Pole
چکیده :
Reinforcement learning algorithms, especially model-free algorithms like Q-learning, have shown reliable results in finding optimal solutions for many real-time applications. However, challenges such as exploration in real-time and the convergence rate need to be addressed, and many researches have proposed algorithms to tackle these challenges. Algorithms like speedy Q-learning, Zap Q-learning, algorithms based on adding a regularization term, noise injection, and many others have been introduced. In this paper, an algorithm based on controlled linear perturbation is presented, which, according to the numerical results, can significantly reduce unnecessary explorations that are risky in real-time. Additionally, the proposed algorithm does not depend on the learning rate \mathbit{\alpha}, \mathbit{\gamma}, or changes in coefficients. However, to be effective, the parameters of the algorithm should be chosen within the correct range. The results of applying the proposed algorithm have been compared with three reliable algorithms: standard Q-learning, speedy Q-learning, and noise injection. These comparisons were conducted in a 9x9 maze scenario and in the cart-pole environment.
لیست مقالات
لیست مقالات بایگانی شده
Enhanced Forward Model for Photoacoustic Imaging with Speed of Sound Compensation
Amirreza Jodeiry - Zahra Kavehvash
Hybrid-Excited, Variable-Flux, and Inter-Modular Biased-Flux Motors: A Comparative Analysis
Mohammad Amirkhani - Ehsan Farmahini Farahani - Alireza Eikani - Mojtaba Mirsalim - Javad Shokrollahi Moghani
A Design Methodology for Submicron Low-Voltage Bandgap Voltage Reference
Mehdi Samavati - Samad Sheikhaei - Mohsen Jalali
Design and Implementation of a Compact 4×4 Microstrip Patch Antenna Array with Circular Polarization for Nano Satellite Applications
Mohammad Ghasemi - Majid Afsahi - Shoresh Namdari
Closed-Form Circularization of Hyperbolic Dispersion with Isotropic Energy Transport
Mohammad Mehdi Sadeghi
A 30dB and 250μW High Linear Variable Gain Amplifier with Employing Gm-boosting and Common Mode Feedforward Techniques
Mehdi Shahabi
A New Optimal Design of a Solar Power Plant On The Rooftop of Bovisa Train Station in Milan
Omid Nasirkhani - Mohsen Tamaddon
Partitioning-based Graph Signal Denoising via Heat Kernel Smoothing
Mohammadreza Fattahi - Hamid Saeedi-Sourck - Vahid Abootalebi
Distributed Data Processing for Multi-Agent Systems Via Wave Model
Saeedreza Tofighi - Masoud Shafiee
Machine Learning-based Fundamental Stock Prediction Using Companies’ Financial Reports
Hossein Rezaei - Kamran Abdi - Mohsen Hooshmand
بیشتر
ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 44.7.2