Key Takeaways
- The new Machine Learning Toolkit from Keysight Technologies reduces model development and extraction time from weeks to hours
- The toolkit features an ML optimizer, auto-extraction flows, and utilities to accelerate parameter extraction and improve predictive accuracy
- The solution supports scalable workflows across multiple technologies, including FinFET, GAA, GaN, SiC, and bipolar devices
- The toolkit enables faster feedback loops between device and circuit design, reducing PDK development cycles from weeks to days
- Keysight’s Machine Learning Toolkit empowers customers to deliver more predictive, higher-quality models in significantly less time, accelerating PDK development and innovation
Introduction to Machine Learning Toolkit
The semiconductor industry is undergoing a rapid transformation, driven by advanced architectures and heterogeneous integration strategies. To address the complex modeling and parameter extraction challenges that come with these innovations, Keysight Technologies has released a new Machine Learning Toolkit. As Nilesh Kamdar, General Manager of Keysight EDA, stated, "AI/ML is fundamentally transforming the traditional workflows and methodologies of compact modeling. With the new Machine Learning Toolkit, we empower our customers to deliver more predictive, higher-quality models in significantly less time — accelerating PDK development and helping them keep pace with rapidly evolving semiconductor technologies." The toolkit reduces model development and extraction time from weeks to hours, enabling faster Process Design Kit (PDK) delivery and Design Technology Co-Optimization (DTCO) applications.
Challenges in Traditional Workflows
Traditional workflows rely on physics-based compact models and manual parameter extraction, which can be a time-consuming and labor-intensive process. Engineers must adjust hundreds of interconnected parameters across multiple operating conditions, a process that can take weeks and often struggles to achieve optimal results. With increasingly tight schedules, faster, more predictive, and automated artificial intelligence/Machine Learning (AI/ML)-driven modeling solutions are now essential. The new Machine Learning Toolkit from Keysight Technologies tackles these challenges by introducing a framework that combines advanced neural network architectures with ML-based optimization. As Kamdar noted, "By leveraging AI/ML-driven modeling, Keysight enables semiconductor companies to accelerate innovation, reduce development risk, and maintain a competitive edge in a rapidly evolving market."
Key Features and Benefits
The Machine Learning Toolkit features an ML optimizer, auto-extraction flows, and utilities to accelerate parameter extraction and improve predictive accuracy. The solution supports scalable workflows across multiple technologies, including FinFET, GAA, GaN, SiC, and bipolar devices. The toolkit enables faster feedback loops between device and circuit design, reducing PDK development cycles from weeks to days. With the new Machine Learning Toolkit, customers can reduce the parameter extraction steps from over 200 to fewer than 10, accelerating PDK delivery, automating DTCO, and speeding up time-to-market. The toolkit also integrates seamlessly with Keysight’s Device Modeling platform, supporting Python-based customization and robust automated modeling flow.
Impact on the Semiconductor Industry
The new Machine Learning Toolkit has the potential to significantly impact the semiconductor industry by accelerating innovation and reducing development risk. By enabling customers to deliver more predictive, higher-quality models in significantly less time, the toolkit can help semiconductor companies maintain a competitive edge in a rapidly evolving market. As Kamdar stated, "With the new Machine Learning Toolkit, we empower our customers to deliver more predictive, higher-quality models in significantly less time — accelerating PDK development and helping them keep pace with rapidly evolving semiconductor technologies." The toolkit can also help reduce the time and cost associated with traditional workflows, making it an attractive solution for companies looking to improve their design and development processes.
Additional Enhancements
In addition to the new Machine Learning Toolkit, Keysight has also introduced several other enhancements across its device modeling solutions. These include new rules related to the Aging Model QA for OMI and MOSRA in Device Modeling MQA 2026, a new remote-control feature for remote low-frequency noise testing with A-LFNA in Device Modeling WaferPro 2025, and a new low-frequency noise stress test capability for seamless measurement from stress test to noise test in A-LFNA 2026. These enhancements demonstrate Keysight’s commitment to providing its customers with the most advanced and innovative solutions for device modeling and parameter extraction.
Conclusion
In conclusion, the new Machine Learning Toolkit from Keysight Technologies is a game-changer for the semiconductor industry. By reducing model development and extraction time from weeks to hours, the toolkit enables faster PDK delivery and DTCO applications, accelerating innovation and reducing development risk. With its scalable workflows, automated modeling flow, and improved predictive accuracy, the toolkit is an attractive solution for companies looking to improve their design and development processes. As Kamdar noted, "By leveraging AI/ML-driven modeling, Keysight enables semiconductor companies to accelerate innovation, reduce development risk, and maintain a competitive edge in a rapidly evolving market." With the new Machine Learning Toolkit, Keysight is empowering its customers to deliver more predictive, higher-quality models in significantly less time, accelerating PDK development and innovation in the semiconductor industry.
https://finance.yahoo.com/news/keysight-unveils-machine-learning-toolkit-160000461.html