<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-21T02:05:53Z</responseDate><request verb="GetRecord" identifier="oai:macsphere.mcmaster.ca:11375/30655" metadataPrefix="dim">https://prod-ms-be.lib.mcmaster.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:macsphere.mcmaster.ca:11375/30655</identifier><datestamp>2026-09-24T03:39:17Z</datestamp><setSpec>com_11375_271</setSpec><setSpec>col_11375_272</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:doc="http://www.lyncode.com/xoai" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">Abdelhadi, Ameer</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author">Taha, Hazem</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="department" lang="en_US">Electrical and Computer Engineering</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2024-12-23T19:50:06Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2024-12-23T19:50:06Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2024</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/11375/30655</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://doi.org/10.71548/10313</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en_US">This thesis presents HEPPO: Hardware-Efficient Proximal Policy Optimization, a framework designed to address the computational and memory challenges associated with implementing advanced reinforcement learning algorithms on resource-constrained hardware platforms. By introducing dynamic standardization for rewards and an 8-bit quantization strategy, HEPPO reduces memory requirements by up to 75% while improving training stability and performance, achieving up to a 67% increase in cumulative rewards. A novel, highly parallelized architecture for Generalized Advantage Estimation (GAE) computation accelerates this critical phase, processing 19.2 billion elements per second using 64 processing elements, contributing to a 22% to 37% reduction in PPO training time in different environments. Adapting the proposed on-chip memory layout reduces the GAE data transfer latency and increases the reduction percentage up to 48% in certain environments in PPO training time. The integration of the entire PPO pipeline on a single System-on-Chip (SoC) further enhances system performance by reducing communication overhead and leveraging custom hardware acceleration. Experimental evaluations demonstrate that HEPPO effectively bridges the gap between sophisticated reinforcement learning algorithms and practical hardware implementations, enabling efficient deployment in embedded systems and real-time applications.</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degreetype" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degree" lang="en_US">Master of Applied Science (MASc)</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="layabstract" lang="en_US">Reinforcement Learning (RL) enables agents to acquire knowledge and make decisions by interacting with their environment, similar to human experiential learning. RL has been widely used in various industrial domains such as health care and finance. However, the implementation of sophisticated RL algorithms on small, resource-limited devices such as embedded systems or edge devices is often difficult due to the fact that they require a significant amount of computational power and memory. This thesis presents HEPPO, a novel framework facilitating the efficient execution of the widely used reinforcement learning algorithm, Proximal Policy Optimization (PPO), across several hardware platforms. By optimizing critical bottlenecks of the algorithm and developing a customized hardware architecture, HEPPO markedly decreases computational requirements and memory consumption without compromising performance. The proposed framework enables the real time deployment of intelligent learning algorithms on devices such as drones, robots, and other smart systems, thereby augmenting their capabilities and facilitating new avenues for innovation across diverse industries.</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en_US">en_US</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Reinforcement Learning (RL)</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Proximal Policy Optimization (PPO)</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Hardware Acceleration</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Generalized Advantage Estimation (GAE)</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Embedded Systems</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Resource-Constrained Environments</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Custom Hardware Architectures</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Quantization Techniques</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">FPGA-Based Acceleration</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Deep Learning Optimization</dim:field>
   <dim:field mdschema="dc" element="title" lang="en_US">HEPPO: Hardware-Efficient Proximal Policy Optimization. A Universal Pipelined Architecture for Generalized Advantage Estimation</dim:field>
   <dim:field mdschema="dc" element="type" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim></metadata></record></GetRecord></OAI-PMH>