Mathematical optimizations for deep learning

Basit Öğe Kaydı
dc.authoridÇetin Kaya Koç / 0000-0002-2572-9565en_US
dc.authorscopusidÇetin Kaya Koç / 57053693300
dc.authorwosidÇetin Kaya Koç / GBX-7437-2022
dc.authorwosidÇetin Kaya Koç / W-3929-2018 wos
dc.contributor.authorGreen, Sam
dc.contributor.authorVineyard, Craig M.
dc.contributor.authorKoç, Çetin Kaya
dc.date.accessioned2020-08-30T20:01:38Z
dc.date.available2020-08-30T20:01:38Z
dc.date.issued2018
dc.departmentİstinye Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Bilgisayar Mühendisliği Bölümüen_US
dc.descriptionKoç, Çetin Kaya (isu author)
dc.description.abstractDeep neural networks are often computationally expensive, during both the training stage and inference stage. Training is always expensive, because back-propagation requires high-precision floating-pointmultiplication and addition. However, various mathematical optimizations may be employed to reduce the computational cost of inference. Optimized inference is important for reducing power consumption and latency and for increasing throughput. This chapter introduces the central approaches for optimizing deep neural network inference: pruning "unnecessary" weights, quantizing weights and inputs, sharing weights between layer units, compressing weights before transferring from main memory, distilling large high-performance models into smaller models, and decomposing convolutional filters to reduce multiply and accumulate operations. In this chapter, using a unified notation, we provide a mathematical and algorithmic description of the aforementioned deep neural network inference optimization methods. © Springer Nature Switzerland AG 2018.en_US
dc.identifier.citationGreen S., Vineyard C.M., Koç Ç.K. (2018) Mathematical Optimizations for Deep Learning. In: Koç Ç.K. (eds) Cyber-Physical Systems Security. Springer, Cham. https://doi.org/10.1007/978-3-319-98935-8_4en_US
dc.identifier.citationGreen, S., Vineyard, C. M., & Koç, Ç. K. (2018). Mathematical Optimizations for Deep Learning. In Cyber-Physical Systems Security (pp. 69-92). Springer, Cham.en_US
dc.identifier.doi10.1007/978-3-319-98935-8_4en_US
dc.identifier.endpage92en_US
dc.identifier.isbn9783319989358; 9783319989341
dc.identifier.scopus2-s2.0-85063927536en_US
dc.identifier.scopusqualityN/Aen_US
dc.identifier.startpage69en_US
dc.identifier.urihttps://doi.org/10.1007/978-3-319-98935-8_4
dc.identifier.urihttps://hdl.handle.net/20.500.12713/330
dc.indekslendigikaynakScopusen_US
dc.institutionauthorKoç, Çetin Kayaen_US
dc.language.isoenen_US
dc.publisherSpringer International Publishingen_US
dc.relation.ispartofCyber-Physical Systems Securityen_US
dc.relation.publicationcategoryKitap Bölümü - Uluslararasıen_US
dc.rightsinfo:eu-repo/semantics/openAccessen_US
dc.titleMathematical optimizations for deep learningen_US
dc.typeBook Chapteren_US

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim
İsim:
10.1007_978-3-319-98935-8.pdf
Boyut:
986.46 KB
Biçim:
Adobe Portable Document Format
Açıklama:
Tam Metin / Full Text
İndir

Koleksiyon

Scopus İndeksli Yayınlar Koleksiyonu
Bilgisayar Mühendisliği Bölümü Kitap Bölümü Koleksiyonu