Deep Reinforcement Learning-based through Silicon Via (TSV) Array Design Optimization Method considering Crosstalk

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 50
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorKim, Keunwooko
dc.contributor.authorPark, Hyunwookko
dc.contributor.authorLho, Daehwanko
dc.contributor.authorKim, Minsuko
dc.contributor.authorSon, Keeyoungko
dc.contributor.authorSon, Kyungjuneko
dc.contributor.authorKim, Seonggukko
dc.contributor.authorShin, Taeinko
dc.contributor.authorChoi, Seongukko
dc.contributor.authorKim, Jounghoko
dc.date.accessioned2023-08-08T12:00:19Z-
dc.date.available2023-08-08T12:00:19Z-
dc.date.created2023-07-07-
dc.date.issued2020-12-
dc.identifier.citation2020 IEEE Electrical Design of Advanced Packaging and Systems, EDAPS 2020-
dc.identifier.issn2151-1225-
dc.identifier.urihttp://hdl.handle.net/10203/311272-
dc.description.abstractIn this paper, we propose the through silicon via (TSV) array design optimization method using deep reinforcement learning (DRL) framework. The agent trained through the proposed method can provide an optimal TSV array that minimizes far-end crosstalk (FEXT) in one single step. We define the state, action, and reward that are elements of the Markov Decision Process (MDP) for optimizing the TSV array considering FEXT and train a deep q network (DQN) agent. For verification, we applied the proposed method to a 3 by 3 through silicon via array at stacked DRAM of High Bandwidth Memory (HBM). The network converged well, and as the result, the proposed method provided the optimal design that satisfies the target FEXT in which 3 dB lower than the initial design.-
dc.languageEnglish-
dc.publisherInstitute of Electrical and Electronics Engineers Inc.-
dc.titleDeep Reinforcement Learning-based through Silicon Via (TSV) Array Design Optimization Method considering Crosstalk-
dc.typeConference-
dc.identifier.scopusid2-s2.0-85099779247-
dc.type.rimsCONF-
dc.citation.publicationname2020 IEEE Electrical Design of Advanced Packaging and Systems, EDAPS 2020-
dc.identifier.conferencecountryCC-
dc.identifier.conferencelocationVirtual-
dc.identifier.doi10.1109/EDAPS50281.2020.9312906-
dc.contributor.localauthorKim, Joungho-
dc.contributor.nonIdAuthorPark, Hyunwook-
dc.contributor.nonIdAuthorLho, Daehwan-
dc.contributor.nonIdAuthorKim, Minsu-
dc.contributor.nonIdAuthorSon, Keeyoung-
dc.contributor.nonIdAuthorSon, Kyungjune-
dc.contributor.nonIdAuthorKim, Seongguk-
dc.contributor.nonIdAuthorShin, Taein-
dc.contributor.nonIdAuthorChoi, Seonguk-
Appears in Collection
EE-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0