<\/p>\n\n\n\n
\ud3ec\uc2a4\ud130\/\ub370\ubaa8 \ubc1c\ud45c\ud300\uc5d0\ub294 \ucc38\uac00\uc0c1\ud488(\ud300\ub2f9 \uc2a4\ud0c0\ubc85\uc2a4 \uae30\ud504\ud2f0\ucf58 2\ub9cc\uc6d0)\uc744 \ub4dc\ub9ac\uace0, \uc6b0\uc218 \ud3ec\uc2a4\ud130\/\ub370\ubaa8 \uc218\uc0c1\ud300\uc5d0\ub294 LG \uc2a4\ud0e0\ubc14\uc774\ubbf8\ub97c 3\ud300\uc5d0\uac8c \ubd80\uc0c1\uc73c\ub85c \ub4dc\ub9b4 \uc608\uc815\uc785\ub2c8\ub2e4.<\/strong><\/p>\n\n\n\n <\/p>\n\n\n\n \ucc38\uac00\ud55c \ud3ec\uc2a4\ud130 \ubaa9\ub85d\uc740 \uc544\ub798\uc640 \uac19\uc2b5\ub2c8\ub2e4. <\/p>\n\n\n <\/p>\n","protected":false},"excerpt":{"rendered":" \ud3ec\uc2a4\ud130\/\ub370\ubaa8 \ubc1c\ud45c\ud300\uc5d0\ub294 \ucc38\uac00\uc0c1\ud488(\ud300\ub2f9 \uc2a4\ud0c0\ubc85\uc2a4 \uae30\ud504\ud2f0\ucf58 2\ub9cc\uc6d0)\uc744 \ub4dc\ub9ac\uace0, \uc6b0\uc218 \ud3ec\uc2a4\ud130\/\ub370\ubaa8 \uc218\uc0c1\ud300\uc5d0\ub294 LG \uc2a4\ud0e0\ubc14\uc774\ubbf8\ub97c 3\ud300\uc5d0\uac8c \ubd80\uc0c1\uc73c\ub85c \ub4dc\ub9b4 \uc608\uc815\uc785\ub2c8\ub2e4. \ucc38\uac00\ud55c \ud3ec\uc2a4\ud130 \ubaa9\ub85d\uc740 \uc544\ub798\uc640 \uac19\uc2b5\ub2c8\ub2e4. \uc18c\uc18d \ubc1c\ud45c\uc790 \uc9c0\ub3c4\uad50\uc218 \uc81c\ubaa9 \uc5f0\uc138\ub300\ud559\uad50 \uc784\ucc44\ubbfc \uae40\uc601\uc11d GuardiaNN: Fast and Secure On-Device Inference in TrustZone Using Embedded SRAM and Cryptographic Hardware \uc5f0\uc138\ub300\ud559\uad50 \ucd5c\uc9c4\uc6b0 \uae40\uc601\uc11d Dataflow Mirroring: Architectural Support for Highly Efficient Fine-Grained Spatial Multitasking Read More …<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":16,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-128","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=\/wp\/v2\/pages\/128","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=128"}],"version-history":[{"count":34,"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=\/wp\/v2\/pages\/128\/revisions"}],"predecessor-version":[{"id":1223,"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=\/wp\/v2\/pages\/128\/revisions\/1223"}],"up":[{"embeddable":true,"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=\/wp\/v2\/pages\/16"}],"wp:attachment":[{"href":"https:\/\/css.or.kr\/csc2023\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=128"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}\n
– \ud559\uc0dd\uad50\ub958\ud68c: 2\uc6d4 6\uc77c (\uc6d4\uc694\uc77c) 18:00 ~ 19:00
– \uc6b0\uc218\ud3ec\uc2a4\ud130 \uc2dc\uc0c1: 2\uc6d4 7\uc77c (\ud654\uc694\uc77c) 17:30 ~ 17:40<\/li>\n\n\n\n\n\n
\n \uc18c\uc18d<\/strong><\/td>\n \ubc1c\ud45c\uc790<\/strong><\/td>\n \uc9c0\ub3c4\uad50\uc218<\/strong><\/td>\n \uc81c\ubaa9<\/strong><\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \uc784\ucc44\ubbfc<\/td>\n \uae40\uc601\uc11d<\/td>\n GuardiaNN: Fast and Secure On-Device Inference in TrustZone Using Embedded SRAM and Cryptographic Hardware<\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \ucd5c\uc9c4\uc6b0<\/td>\n \uae40\uc601\uc11d<\/td>\n Dataflow Mirroring: Architectural Support for Highly Efficient Fine-Grained Spatial Multitasking on Systolic-Array NPUs<\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \uc774\uc815\ud6c4<\/td>\n \uae40\uc601\uc11d<\/td>\n GCoM: A Detailed GPU Core Model for Accurate Analytical Modeling of Modern GPUs<\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \uc774\uc6a9\uc6b0<\/td>\n \uae40\ud55c\uc900<\/td>\n HECATE: Performance-Aware Scale Optimization for Homomorphic Encryption Compiler<\/td>\n<\/tr>\n \n \uad6d\ubbfc\ub300\ud559\uad50<\/td>\n \uc548\uc194<\/td>\n \uae40\uc885\ucc2c<\/td>\n Demand Layering for Real-Time DNN Inference with Minimized Memory Usage (Demo)<\/td>\n<\/tr>\n \n \ud56d\uacf5\ub300\ud559\uad50<\/td>\n \uc774\uc218\uad11<\/td>\n \uc774\uc7ac\ud658<\/td>\n \ucee8\ud14c\uc774\ub108 \ud658\uacbd\uc5d0\uc11c\uc758 PyTorch \ubd84\uc0b0 \ub525\ub7ec\ub2dd \ud1b5\uc2e0 \ub77c\uc774\ube0c\ub7ec\ub9ac\uc758 \uc2e4\uc99d\uc801 \uc131\ub2a5 \ud3c9\uac00<\/td>\n<\/tr>\n \n DGIST<\/td>\n \uac15\uc6b0\uc131<\/td>\n \uc88c\ud6c8\uc2b9<\/td>\n DNN-SAM: Split-and-Merge DNN Execution for Real-Time Object Detection<\/td>\n<\/tr>\n \n KAIST<\/td>\n \ud669\uc9c4\uc6b0<\/td>\n \ubc15\uc885\uc138<\/td>\n CoVA: Exploiting Compressed-Domain Analysis to Accelerate Video Analytics<\/td>\n<\/tr>\n \n POSTECH<\/td>\n \ud568\ud615\uaddc<\/td>\n \uae40\uad11\uc120<\/td>\n The Case for Memory Access-Triggered Near-Data Processing for Accelerating DNN Training on GPUs<\/td>\n<\/tr>\n \n \uc11c\uc6b8\ub300\ud559\uad50<\/td>\n \uc774\uc720\ucc2c<\/td>\n \uc774\ubcd1\uc601<\/td>\n LegoExp: Automated Kernel Information Leakage Exploit Generation for Heap Temporal Memory Vulnerability<\/td>\n<\/tr>\n \n \uacbd\ubd81\ub300\ud559\uad50<\/td>\n \uae40\uc720\uc591<\/td>\n \ud0c1\ubcd1\ucca0<\/td>\n \uc628\ub77c\uc778 \ud3ec\ub7fc \uc9c0\uc2dd \ud65c\uc6a9\uc744 \ud1b5\ud55c \ubb38\uc81c \ud574\uacb0 \ub2a5\ub825 \ud5a5\uc0c1<\/td>\n<\/tr>\n \n \uc11c\uc6b8\ub300\ud559\uad50<\/td>\n \uc1a1\uc218\ud658<\/td>\n \uc774\ubcd1\uc601<\/td>\n Fuzzing on Rendering Engines to Find Render-update Bugs<\/td>\n<\/tr>\n \n \ud56d\uacf5\ub300\ud559\uad50<\/td>\n \uc7a5\uc218\ud638<\/td>\n \uc774\uc7ac\ud658<\/td>\n GPGPU\ub97c \uc774\uc6a9\ud55c local node connectivity(node independent paths) \uacc4\uc0b0 \uac00\uc18d\ud654<\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \uc774\uc7ac\ud638<\/td>\n \uae40\ud55c\uc900<\/td>\n Memory-efficient GPU Compiler for DNN Inference<\/td>\n<\/tr>\n \n \uad6d\ubbfc\ub300\ud559\uad50<\/td>\n \uc548\ud0dc\uc6b1<\/td>\n \uae40\uc885\ucc2c<\/td>\n Scale Truck Platooning Research Testbed (Demo)<\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \uc774\uc6a9\uc6b0<\/td>\n \uae40\ud55c\uc900<\/td>\n ELASM: Error-Latency-Aware Scale Management for Fully Homomorphic Encryption<\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \uc815\uc2e0\ub155<\/td>\n \uae40\ud55c\uc900<\/td>\n Decoupling scheduler, topology layout, and algorithm to easily enlarge the tuning space of gpu graph processing<\/td>\n<\/tr>\n \n \uc11c\uac15\ub300\ud559\uad50<\/td>\n \ubc15\uc5ec\ud604<\/td>\n \uae40\uc601\uc7ac<\/td>\n Background Task-Aware Scheduling in SPDK<\/td>\n<\/tr>\n \n \uc11c\uc6b8\ub300\ud559\uad50<\/td>\n \uae40\uc0c1\uc724<\/td>\n \uc774\ubcd1\uc601<\/td>\n Device only attack with memory corruption without user interaction<\/td>\n<\/tr>\n \n \uc131\uade0\uad00\ub300\ud559\uad50<\/td>\n \uc11c\uc885\uc120<\/td>\n \uc870\ud615\ubbfc<\/td>\n Test program generation for processer core verification using reinforcement learning<\/td>\n<\/tr>\n \n \uc131\uade0\uad00\ub300\ud559\uad50<\/td>\n \uc548\uc601\ud0dc \/ \uae40\uc7ac\ud615<\/td>\n \uc870\ud615\ubbfc<\/td>\n SEAL-Compatible Fully Homomorphic Encryption Accelerator<\/td>\n<\/tr>\n \n \uc544\uc8fc\ub300\ud559\uad50<\/td>\n \uc815\uc9c4\uc6b0<\/td>\n \uc548\uc815\uc12d<\/td>\n Efficient Deep Learning Model Serving with Direct-Host-Access<\/td>\n<\/tr>\n \n \uace0\ub824\ub300\ud559\uad50<\/td>\n \ubc15\ubcf4\uc601<\/td>\n \uad6c\uac74\uc7ac<\/td>\n SSD\uc758 \ub0b4\ubd80 \uc5f0\uc0b0\uc744 \uc774\uc6a9\ud55c \uc720\uc804\uc790 \ubd84\uc11d \uc5b4\ud50c\ub9ac\ucf00\uc774\uc158 \uac00\uc18d<\/td>\n<\/tr>\n \n \uc5f0\uc138\ub300\ud559\uad50<\/td>\n \uc784\ucc44\ubbfc<\/td>\n \uae40\uc601\uc11d<\/td>\n Design and Analysis of a Processing-in-DIMM Join Algorithm: A Case Study with UPMEM DIMMs<\/td>\n<\/tr>\n \n \uace0\ub824\ub300\ud559\uad50<\/td>\n \uc815\uc885\ud604<\/td>\n \uad6c\uac74\uc7ac<\/td>\n An Efficient Memory Controller Scheme for SIMT Architecture<\/td>\n<\/tr>\n \n \uc131\uade0\uad00\ub300\ud559\uad50<\/td>\n \ubc15\ubbfc\uac74<\/td>\n \uc870\ud615\ubbfc<\/td>\n MiRU : Accelerating Depthwise Convolutions via Maximizing Input Reuse on Systolic Arrays<\/td>\n<\/tr>\n \n \uc11c\uac15\ub300\ud559\uad50<\/td>\n \uc774\uc11c\uc601<\/td>\n \uae40\uc601\uc7ac<\/td>\n Fully Asynchronous Data Deduplication Using Write-Ahead Log in LSM-tree based Key-Value Stores<\/td>\n<\/tr>\n \n POSTECH<\/td>\n \ud64d\uc815\ubbfc<\/td>\n \uae40\uad11\uc120<\/td>\n Overcoming GPU Memory Capacity Wall with 3D-stacked DRAM and Storage Class Memory<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n