dc.contributor.advisor | Hossain, Muhammad Iqbal | |
dc.contributor.advisor | Mukta, Jannatun Noor | |
dc.contributor.author | Bhuiyan, Emtiaz MD Tafsir | |
dc.contributor.author | Rahman, Mushfiqur | |
dc.contributor.author | Mondal, Sudipta | |
dc.contributor.author | Warech, Sadman | |
dc.date.accessioned | 2021-09-07T10:03:32Z | |
dc.date.available | 2021-09-07T10:03:32Z | |
dc.date.copyright | 2021 | |
dc.date.issued | 2021-06 | |
dc.identifier.other | ID 16301049 | |
dc.identifier.other | ID 20241040 | |
dc.identifier.other | ID 17301224 | |
dc.identifier.other | ID 16301115 | |
dc.identifier.uri | http://hdl.handle.net/10361/14982 | |
dc.description | This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2021. | en_US |
dc.description | Cataloged from PDF version of thesis. | |
dc.description | Includes bibliographical references (pages 53-56). | |
dc.description.abstract | It’s challenging to provide security for cloud-based services, especially for cloud
processing services, due to the fact that typical encryption techniques do not al low for calculation on encrypted data. The formation of Homomorphic Encryption
techniques shows significant possibilities of incorporating encrypted computation on
cloud infrastructures. This enables owners to outsource computation over confiden tial data to cloud vendors. Control and synthesis tasks of sensitive systems like
traffic light control, article recommendation for online users and potentially, robot’s
action determination can be delegated to a cloud-based Reinforcement Learning
agent. In this study, we designed two Deep Reinforcement Learning agents that
work on ciphertexts using Homomorphic Encryption. Both agents take encrypted
state images and produce encrypted actions. One learns on plain data but evaluates
on encrypted inputs, while the other one operates fully on encrypted space. The
performance of both agents is compared against plaintext RL agents with identical
parameters. The paper also describes possible architectures for such systems. | en_US |
dc.description.statementofresponsibility | Emtiaz MD Tafsir Bhuiyan | |
dc.description.statementofresponsibility | Mushfiqur Rahman | |
dc.description.statementofresponsibility | Sudipta Mondal | |
dc.description.statementofresponsibility | Sadman Warech | |
dc.format.extent | 56 pages | |
dc.language.iso | en | en_US |
dc.publisher | Brac University | en_US |
dc.rights | Brac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. | |
dc.subject | Homomorphic Encryption | en_US |
dc.subject | Privacy preserving | en_US |
dc.subject | Reinforcement Learning | en_US |
dc.subject | Deep Q Learning | en_US |
dc.subject.lcsh | Reinforcement Learning | |
dc.title | Implementation of real-time learning on homomorphically encrypted visual inputs | en_US |
dc.type | Thesis | en_US |
dc.contributor.department | Department of Computer Science and Engineering, Brac University | |
dc.description.degree | B. Computer Science | |