Topology-oriented pattern extraction and classification for synthesizing lithography test patterns

Shim, Seongbo; Shin, Youngsoo

doi:10.1117/1.jmm.14.1.013503

Cited by 18 publications

(8 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, fuzzy matching fails to extract problematic instances on a more difficult testcase ICCAD16-4 with looser constraints that they all reach less than 50% prediction accuracy. [3,13] propose to use the frequency domain representation to sample layout patterns with similar property and detect hotspots. Here we conduct additional experiments by clustering layout clips based on their Fourier Transform results.…”

Section: Comparison With Existing Methodsmentioning

confidence: 99%

“…Layout pattern sampling problems are addressed by several works that are, to some extent, related to clustering approaches. Representative methods include clustering on frequency domain [3,13], Bayesian clustering [14], and clustering based on layout topology [9,10,13,15]. However, sampling and hotspot detection are mostly conducted exclusively which ignores the beneath integrity between them.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bridging the Gap Between Layout Pattern Sampling and Hotspot Detection via Batch Active Learning

Yang

Tabery³

et al. 2018

Preprint

View full text Add to dashboard Cite

Layout hotpot detection is one of the main steps in modern VLSI design. A typical hotspot detection flow is extremely time consuming due to the computationally expensive mask optimization and lithographic simulation. Recent researches try to facilitate the procedure with a reduced flow including feature extraction, training set generation and hotspot detection, where feature extraction methods and hotspot detection engines are deeply studied. However, the performance of hotspot detectors relies highly on the quality of reference layout libraries which are costly to obtain and usually predetermined or randomly sampled in previous works. In this paper, we propose an active learning-based layout pattern sampling and hotspot detection flow, which simultaneously optimizes the machine learning model and the training set that aims to achieve similar or better hotspot detection performance with much smaller number of training instances. Experimental results show that our proposed method can significantly reduce lithography simulation overhead while attaining satisfactory detection accuracy on designs under both DUV and EUV lithography technologies.

show abstract

Section: Comparison With Existing Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Bridging the Gap Between Layout Pattern Sampling and Hotspot Detection via Batch Active Learning

Yang

Tabery³

et al. 2018

Preprint

View full text Add to dashboard Cite

show abstract

“…Instead, a ML model should properly leverage both local high frequency and global low frequency components of a mask image for high lithography simula-tion accuracy. Furthermore, spectral analysis has been studied on layout-related applications [15,16], showing that frequency domain representations can more powerfully reflect global layout characteristics.…”

Section: Mask Imagementioning

confidence: 99%

Generic Lithography Modeling with Dual-band Optics-Inspired Neural Networks

Heng¹,

Li²,

Sastry³

et al. 2022

Preprint

View full text Add to dashboard Cite

Lithography simulation is a critical step in VLSI design and optimization for manufacturability. Existing solutions for highly accurate lithography simulation with rigorous models are computationally expensive and slow, even when equipped with various approximation techniques. Recently, machine learning has provided alternative solutions for lithography simulation tasks such as coarse-grained edge placement error regression and complete contour prediction. However, the impact of these learningbased methods has been limited due to restrictive usage scenarios or low simulation accuracy. To tackle these concerns, we introduce an dual-band optics-inspired neural network design that considers the optical physics underlying lithography. To the best of our knowledge, our approach yields the first published via/metal layer contour simulation at 1nm 2 /pixel resolution with any tile size. Compared to previous machine learning based solutions, we demonstrate that our framework can be trained much faster and offers a significant improvement on efficiency and image quality with 20× smaller model size. We also achieve 85× simulation speedup over traditional lithography simulator with ∼ 1% accuracy loss.

show abstract

“…Actual patterns are extracted from sample layouts and can cover more random shapes, but some similar shapes may be more frequent and some are not really important. Thus the extraction of important shapes, e.g., extracting hotspot patterns [26], and classifying them are important.…”

Section: Test Patternsmentioning

confidence: 99%

“…These strict pattern matching methods do not capture the similarity when one pattern is a shifted (or rotated or reflected) version of the other. This can be alleviated through pattern matching in Fourier domain [26]; the two patterns in Fig. 14 (a), which are similar by just 10%, are now very similar in frequency domain as illustrated in Fig.…”

Section: Classificationmentioning

confidence: 99%

Computational Lithography Using Machine Learning Models

Shin

2021

IPSJ Transactions on System LSI Design Methodology

Self Cite

View full text Add to dashboard Cite

Machine learning models have been applied to a wide range of computational lithography applications since around 2010. They provide higher modeling capability, so their application allows modeling of higher accuracy. Many applications which are computationally expensive can take advantage of machine learning models, since a well trained model provides a quick estimation of outcome. This tutorial reviews a number of such computational lithography applications that have been using machine learning models. They include mask optimization with OPC (optical proximity correction) and EPC (etch proximity correction), assist features insertion and their printability check, lithography modeling with optical model and resist model, test patterns, and hotspot detection and correction.

show abstract

Topology-oriented pattern extraction and classification for synthesizing lithography test patterns

Cited by 18 publications

References 20 publications

Bridging the Gap Between Layout Pattern Sampling and Hotspot Detection via Batch Active Learning

Bridging the Gap Between Layout Pattern Sampling and Hotspot Detection via Batch Active Learning

Generic Lithography Modeling with Dual-band Optics-Inspired Neural Networks

Computational Lithography Using Machine Learning Models

Contact Info

Product

Resources

About