Výsledky vyhledávání

Using Deep Neural Networks And Derivative Free Optimization To Accelerate Coverage Closure

Autor: Eldad Haber, Brian Irwin, Raviv Gal, Bilal Saleh, Marwa Mouallem, Avi Ziv

Publikováno v: MLCAD

In computer aided design (CAD), a core task is to optimize the parameters of noisy simulations. Derivative free optimization (DFO) methods are the most common choice for this task. In this paper, we show how four DFO methods, specifically implicit fi

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::3577bef830fe1e0987ea1895c2e9d5f8
https://doi.org/10.1109/mlcad52597.2021.9531234

Zobrazit plný text záznamu

Automatic Scalable System for the Coverage-Directed Generation (CDG) Problem

Autor: Ziv Nevo, Brian Irwin, Wesam Ibraheem, Raviv Gal, Eldad Haber, Avi Ziv

Publikováno v: DATE

We present AS-CDG, a novel automatic scalable system for data-driven coverage-directed generation. The goal of AS-CDG is to find the test templates that maximize the probability of hitting uncovered events. The system contains two phases, one for a c

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::86ad5ca51430cd503c85c4c1ea6eaacc
https://doi.org/10.23919/date51398.2021.9474160

Zobrazit plný text záznamu

Using Machine Learning Clustering To Find Large Coverage Holes

Autor: Giora Simchoni, Raviv Gal, Avi Ziv

Publikováno v: MLCAD

Identifying large and important coverage holes is a time-consuming process that requires expertise in the design and its verification environment. This paper describes a novel machine learning-based technique for finding large coverage holes when the

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::045c9ca0407ebe69c02eed2ba77910c1
https://doi.org/10.1145/3380446.3430621

Zobrazit plný text záznamu

Using DNNs and Smart Sampling for Coverage Closure Acceleration

Autor: Eldad Haber, Avi Ziv, Raviv Gal

Publikováno v: MLCAD

Coverage Directed Generation represents algorithms that are used to create tests or test-templates for hitting coverage events. Standard approaches for solving the problem use either user's intuition or random sampling. Recent work has been using opt

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::54262eb6e8138d64fd430bd77d96cc51
https://doi.org/10.1145/3380446.3430627

Zobrazit plný text záznamu

ML for CAD - Where is the Treasure Hiding?

Autor: Manish Jatashanker Pandey, David Z. Pan, Raviv Gal, Avi Ziv, Haoxing Ren, Marilyn Wolf

Publikováno v: MLCAD

Advances in ML have revolutionized its effectiveness for a variety of applications. Indeed, in areas like image classification and NLP, ML (AI) has changed the rules of the game and opened the door to incredible advances. Design processes seem to mat

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::dcdd20becc59c27b30018a3c03b95c8e
https://doi.org/10.1145/3380446.3431818

Zobrazit plný text záznamu

Late Breaking Results: FRIENDS - Finding Related Interesting Events via Neighbor Detection

Autor: Raviv Gal, Avi Ziv, Haim Kermany, Alexander Ivrii, Ziv Nevo

Publikováno v: DAC

We present Finding Related Interesting Events via Neighbor Detection (FRIENDS), a novel approach to assist verification teams with coverage closure. FRIENDS uses formal verification to find the neighboring events of a never-hit, or hard-to-hit, cover

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::c3c5ad6de049ca8c4227a4d1240f44c8
https://doi.org/10.1109/dac18072.2020.9218685

Zobrazit plný text záznamu

Challenges and Solutions in Post-Silicon Validation of High-end Processors (Invited Tutorial)

Autor: Avi Ziv

Publikováno v: FMCAD

Due to the complexity of designs, post-silicon validation remains a major challenge with few systematic solutions. In this tutorial, we describe the main challenges in post-silicon validation of high-end processors and provide an overview of the stat

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::7d23014734bad1d3d97fa7c8d1d1bc1e
https://doi.org/10.23919/fmcad.2019.8894258

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání