BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000611Z
LOCATION:704-706
DTSTART;TZID=America/Denver:20231113T140000
DTEND;TZID=America/Denver:20231113T173000
UID:submissions.supercomputing.org_SC23_sess457@linklings.com
SUMMARY:Workshop on Machine Learning with Graphs in High Performance Compu
 ting Environments
DESCRIPTION:Addressing Stale Gradients in Scalable Federated Deep Reinforc
 ement Learning\n\nAdvancements in reinforcement learning (RL) via deep neu
 ral networks have enabled their application to a variety of real-world pro
 blems. However, these applications often suffer from long training times. 
 While attempts to distribute training have been successful in controlled s
 cenarios, they face ch...\n\n\nJustin Stanley and Ali Jannesari (Iowa Stat
 e University)\n---------------------\nAn Analysis of Graph Neural Network 
 Memory Access Patterns\n\nGraph Neural Networks (GNNs) are becoming increa
 singly popular for applying neural networks to graph data. However, as the
  size of the input graph increases, the GPU memory wall problem becomes an
  important issue. Since both current solutions to reduce the memory footpr
 int, such as mini-batch approa...\n\n\nSatoshi Iwata (University of Wiscon
 sin, Fujitsu Ltd); Remzi H. Arpaci-Dusseau (University of Wisconsin); and 
 Akihiko Kasagi (Fujitsu Ltd)\n---------------------\nHPC-GPT: Integrating 
 Large Language Model for High-Performance Computing\n\nLarge Language Mode
 ls (LLMs), including the LLaMA model, have exhibited their efficacy across
  various general-domain natural language processing (NLP) tasks. However, 
 their performance in high-performance computing (HPC) domain tasks has bee
 n less than optimal due to the specialized expertise requir...\n\n\nXianzh
 ong Ding (University of California, Merced); Le Chen (Iowa State Universit
 y); Murali Emani (Argonne National Laboratory (ANL)); Chunhua Liao, Pei-Hu
 ng Lin, and Tristan Vanderbruggen (Lawrence Livermore National Laboratory 
 (LLNL)); Zhen Xie (Argonne National Laboratory (ANL)); and Alberto Cerpa a
 nd Wan Du (University of California, Merced)\n---------------------\nInvit
 ed Talk:  Practical Machine Learning on Biological Knowledge Graphs\n\nJus
 tin Reese (Lawrence Berkeley National Laboratory (LBNL))\n----------------
 -----\nMLG-HPCE – Afternoon Break\n---------------------\nWelcome Machine 
 Learning with Graphs in High Performance Computing Environment\n\nSeung-Hw
 an Lim (Oak Ridge National Laboratory (ORNL))\n---------------------\nAn E
 fficient Distributed Graph Engine for Deep Learning on Graphs\n\nTradition
 al graph-processing algorithms have been widely used in Graph Neural Netwo
 rks (GNNs). Current approaches to graph processing in deep learning face t
 wo main problems. Firstly, easy-to-use deep learning libraries lack suppor
 t for widely used graph processing algorithms and do not provide low-...\n
 \n\nGangda Deng, Ömer Akgül, and Hongkuan Zhou (University of Southern Cal
 ifornia (USC)); Hanqing Zeng, Yinglong Xia, and Jianbo Li (Meta); and Vikt
 or Prasanna (University of Southern California (USC))\n-------------------
 --\nDDStore:  Distributed Data Store for Scalable Training of Graph Neural
  Networks on Large Atomistic Modeling Datasets\n\nGraph neural networks (G
 NNs) are a class of Deep Learning models used in designing atomistic mater
 ials for effective screening of large chemical spaces. To ensure robust pr
 ediction, GNN models must be trained on large volumes of atomistic modelin
 g data on leadership class supercomputing facilities. ...\n\n\nJong Youl C
 hoi, Massimiliano Lupo Pasini, Pei Zhang, Kshitij Mehta, and Frank Liu (Oa
 k Ridge National Laboratory (ORNL)) and Jonghyun Bae and Khaled Ibrahim (L
 awrence Berkeley National Laboratory (LBNL))\n\nTag: Artificial Intelligen
 ce/Machine Learning, Graph Algorithms and Frameworks\n\nRegistration Categ
 ory: Workshop Reg Pass\n\nSession Chairs: Seung-Hwan Lim (Oak Ridge Nation
 al Laboratory (ORNL)); José Moreira (IBM); Catherine Schuman (University o
 f Tennessee, Knoxville); and Richard Vuduc (Georgia Institute of Technolog
 y)
END:VEVENT
END:VCALENDAR