BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250822T115805Z
LOCATION:Room 5.2D02
DTSTART;TZID=Europe/Stockholm:20250617T150000
DTEND;TZID=Europe/Stockholm:20250617T170000
UID:submissions.pasc-conference.org_PASC25_sess129@linklings.com
SUMMARY:MS4F - Machine Learning Support for the Lifetime of Software (ML4S
 W)
DESCRIPTION:Software plays a critical role in scientific discovery across 
 computational science domains, including chemistry, climate science, physi
 cs, and applied mathematics. As software development advances, Machine Lea
 rning (ML) is becoming an essential tool for enhancing developer productiv
 ity, optimizing application execution, and replacing computationally expen
 sive simulations with surrogate Neural Network models. However, several ch
 allenges hinder the broad adoption of ML in software, particularly in the 
 context of sustainable development. With increasingly complex software sta
 cks, workflows, and heterogeneous systems, novel techniques are needed to 
 support development, execution orchestration, and performance optimization
 . A promising approach for reducing software development overhead in High 
 Performance Computing (HPC) is program synthesis, where software is automa
 tically generated from high-level specifications. Large Language Models (L
 LMs) such as GPT-4, Code Llama, and StarCoder provide intelligent code gen
 eration capabilities, yet challenges related to correctness, verification,
  and reliability remain. Understanding these limitations is crucial for im
 proving ML-driven software development. The ML4SW minisymposium serves as 
 a platform for researchers, developers, and industry professionals to expl
 ore ML-driven software synthesis, correctness verification, and applicatio
 n optimization. Key discussions will address ML’s role in enhancing softwa
 re development, ensuring trustworthiness, and integrating ML into real-wor
 ld applications for sustainable and efficient computing.\n\nDesign and Use
  of Energy-Efficient Systems in the Deep Learning Era\n\nModern GPUs, toge
 ther with larger datasets, facilitate the exponential growth and adoption 
 of deep learning models. The training and deployment of deep neural networ
 ks in widely used large-scale data centers, on the other hand, exhibit low
  GPU hardware utilization, barely reaching 50%, as shown by s...\n\n\nPame
 la Delgado (HES-SO)\n---------------------\nLeveraging AI-Driven Code Gene
 ration for Portable and Scalable Simulations\n\nAs high performance comput
 ing (HPC) applications continue to push the boundaries of complexity and p
 erformance, leveraging AI-driven code generation has emerged as a powerful
  approach to streamline scientific software development processes.\nThis t
 alk explores the role of AI code generation models as...\n\n\nOsman Seckin
  Simsek and Florina Ciorba (University of Basel)\n---------------------\nF
 rom Reactive Debugging to Proactive Detection: ML for Performance-Aware So
 ftware Development\n\nSoftware performance evolves over time, yet traditio
 nal debugging and profiling remain reactive, costly, and disconnected from
  development workflows. Performance drift—gradual degradation in execution
  efficiency due to code modifications—often goes undetected until it cause
 s significant ...\n\n\nTanzima Islam (Texas State University)\n-----------
 ----------\nTADASHI: Enabling ML with Correct Code Transformations\n\nAs t
 he landscape of machine learning (ML) continues to evolve, the integration
  of generative AI has become a focal point for automating code generation.
  While it is perfectly suitable to generate text for humans such as the ab
 stract you're reading, this approach often falls short in ensuring the cor
 ...\n\n\nEmil Vatai and Aleksandr Drozd (RIKEN); Ivan R. Ivanov (Institute
  of Science Tokyo, RIKEN); Joao E. Batista (RIKEN); Yinghao Ren (SenseTime
  Research and PowerTensors.AI); and Mohamed Wahib (RIKEN)\n\nDomain: Compu
 tational Methods and Applied Mathematics\n\nSession Chair: Florina Ciorba 
 (University of Basel)
END:VEVENT
END:VCALENDAR