Scholarlink

Design Space Exploration (DSE) in Very Large-Scale Integration (VLSI) is critical for optimizing power and performance in semiconductor design. Traditional manual approaches are impractical due to increasing circuit complexity. Integrating Machine Learning (ML) into DSE automates exploration, predicting power consumption based on configurations. ML-driven DSE reduces search time, leveraging reinforcement learning for iterative refinement. ML reveals non-intuitive design-power relationships, aiding discovery of energy-efficient architectures. This synergy empowers efficient trade-off navigation, advancing VLSI towards low-power, high-performance solutions.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Design Space Exploration in VLSI Using Machine Learning for Power and Performance Optimization

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • B. Satyanarayana,
  • M. Ravi,
  • Pokala Krishnaiah,
  • Chilukuri Dileep,
  • B. Annapoorna,
  • M. Janga Reddy

摘要

Design Space Exploration (DSE) in Very Large-Scale Integration (VLSI) is critical for optimizing power and performance in semiconductor design. Traditional manual approaches are impractical due to increasing circuit complexity. Integrating Machine Learning (ML) into DSE automates exploration, predicting power consumption based on configurations. ML-driven DSE reduces search time, leveraging reinforcement learning for iterative refinement. ML reveals non-intuitive design-power relationships, aiding discovery of energy-efficient architectures. This synergy empowers efficient trade-off navigation, advancing VLSI towards low-power, high-performance solutions.