Please use this identifier to cite or link to this item:
https://hdl.handle.net/20.500.14279/33101
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Acosta, Jorge Sierra | - |
dc.contributor.author | Diavastos, Andreas | - |
dc.contributor.author | Gonzalez, Antonio | - |
dc.date.accessioned | 2024-10-15T07:40:33Z | - |
dc.date.available | 2024-10-15T07:40:33Z | - |
dc.date.issued | 2022 | - |
dc.identifier.citation | IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2022, 22-24 May | en_US |
dc.identifier.isbn | 9781665459549 | - |
dc.identifier.uri | https://hdl.handle.net/20.500.14279/33101 | - |
dc.description.abstract | In this work, we extend the auto-tuning process of the state-of-the-art TVM framework with XFeatur; a tool that extracts new meaningful hardware-related features that improve the quality of the representation of the search space and consequently improve the accuracy of its prediction algorithm. These new features provide information about the amount of thread-level parallelism, shared memory usage, register usage, dynamic instruction count and memory access dependencies. Optimizing ResNet-18 with the proposed features improves the quality of the search space representation by 63% on average and a maximum of 2× for certain tasks, while it reduces the tuning time by 9% (approximately 1.1 hours) and produces configurations that have equal or better performance (up to 92.7%) than the baseline. | en_US |
dc.format | en_US | |
dc.language.iso | en | en_US |
dc.rights | © IEEE | en_US |
dc.title | XFeatur: Hardware Feature Extraction for DNN Auto-tuning | en_US |
dc.type | Conference Papers | en_US |
dc.collaboration | Universitat Politècnica de Catalunya | en_US |
dc.subject.category | Computer and Information Sciences | en_US |
dc.country | Spain | en_US |
dc.subject.field | Engineering and Technology | en_US |
dc.relation.conference | IEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2022 | en_US |
dc.identifier.doi | 10.1109/ISPASS55109.2022.00013 | en_US |
dc.identifier.scopus | 2-s2.0-85134330899 | - |
dc.identifier.url | https://api.elsevier.com/content/abstract/scopus_id/85134330899 | - |
cut.common.academicyear | 2022-2023 | en_US |
item.fulltext | No Fulltext | - |
item.languageiso639-1 | en | - |
item.grantfulltext | none | - |
item.openairecristype | http://purl.org/coar/resource_type/c_c94f | - |
item.cerifentitytype | Publications | - |
item.openairetype | conferenceObject | - |
crisitem.author.dept | Department of Electrical Engineering, Computer Engineering and Informatics | - |
crisitem.author.faculty | Faculty of Engineering and Technology | - |
crisitem.author.orcid | 0000-0002-7139-4444 | - |
crisitem.author.parentorg | Faculty of Engineering and Technology | - |
Appears in Collections: | Δημοσιεύσεις σε συνέδρια /Conference papers or poster or presentation |
CORE Recommender
Items in KTISIS are protected by copyright, with all rights reserved, unless otherwise indicated.