Our research work “Applying Machine Learning Models on Metrology Data for Predicting Device Electrical Performance" has been accepted at AI for Manufacturing Workshop co-located with ECML-PKDD 2023.

Short summary: Moore’s Law states that transistor density will double every two years, which is sustained until today due to continuous innovations, such as multi-patterning. The most important metric to evaluate the quality of printed patterns is edge placement error (EPE). Overlay errors are its largest contribution and can lead to fatal failures, making essential to develop effective analysis and control techniques. The goals of this research work are: (a) to quantify the impact of overlay and (b) to predict at an early stage the final capacitance measurements comparing various ML models on overlay data collected from imec N-14 BEOL process flow, at different process steps (litho, etch, Chemical-Mechanical-Polishing etc.). We demonstrate that with appropriate ML models we are able to do better prediction of electrical results.

Congratulations to all AP-ML team members Sara Sacchi, Victor Blanco Carballo, Philippe Leray. Congratulations to Halder Sandip and Tuan Ngo (University of Barcelona).

2 of our research works have been accepted at SPIE Photomask Technology + EUV Lithography conference, Monterey, California, United States.

1. “Deep learning denoiser assisted roughness measurements extraction from thin resists with low signal-to-noise-ratio (SNR) SEM Images: analysis with SMILE"

2. “Single exposure EUV process optimization for SNLP and BLP layer for next-generation DRAM manufacturing“ (*in collaboration with SCREEN SPE Germany GmbH (Belgium))

Congratulations to all AP-ML team members Sara Sacchi, Victor Blanco Carballo. Congratulations to Halder Sandip. 

Two of our recent research manuscripts/works have been accepted to be presented at 60th Design Automation Conference (DAC) 2023.

1. "A Deep Learning Framework for Verilog Autocompletion Towards Design and Verification Automation" (as a poster at Work-in-Progress (WIP) session)

2. “A Machine Learning Approach Towards SKILL Code Autocompletion” (as a poster at 60 DAC Engineering Front-End Design Track )

** Both are our Work-In-Progress researches.... 

Congratulations to all co-authors Enrique Dehaerne, Halder Sandip, Stefan De Gendt. 

3 of our research works have been accepted at 38th European Mask and Lithography Conference (EMLC) 2023 Dresden, Germany.

1. “SEMI-CenterNet: A Machine Learning Facilitated Approach for Semiconductor Defect Inspection"

2. “YOLOv8 for defect inspection of hexagonal directed self- assembly patterns: a data-centric approach“

3.  “A Deep Learning Facilitated Approach for SEM Image Denoising Towards Improved Contour Detection for DRAM (SNLP+BLP) 2D-Structures” (*in collaboration with Siemens EDA, (U.S.A.))

Congratulations to all AP-ML team members Vic De Ridder, Enrique Dehaerne, Halder Sandip 

2 of our research works have been accepted at 65th International Symposium ELMAR-2023, Zadar, Croatia.

1. “Benchmarking Feature Extractors for Reinforcement Learning-Based Semiconductor Defect Localization"

2. “SEMI-DiffusionInst: A Diffusion Model Based Approach for Semiconductor Defect Classification and Segmentation“

Congratulations to all AP-ML team members Vic De Ridder, Enrique Dehaerne, Halder Sandip, Stefan De Gendt 

Four of our recent research works have been accepted at SPIE Advanced Lithography + Patterning 2023 conference [Optical and EUV Nanolithography XXXVI/Metrology, Inspection, and Process Control XXXVII].

1. “SEMI-PointRend: Improved semiconductor wafer defect classification and segmentation as rendering”

2. "Unsupervised deep learning approach for voltage contrast (VC) image denoising towards device pillars yield analysis"

3."Optimizing YOLOv7 for semiconductor defect detection"

4. "Meander fork defectivity through EUV SADP process from P21 to P26: Relax or tighten design rules?"

Congrats to our APML research team members: Bappaditya Dey, Enrique Dehaerne, Halder Sandip, MinJin Hwang, Victor Blanco, Yannick Hermans. 

It is a great pleasure to share that our research blogpost "Ensemble Deep Learning-based Defect Classification and Detection in SEM Images" has been announced TOP 3 blogpost entries in LearnOpenCV Blog Olympics 2021.

We have been placed 3rd among 1300+ entries and awarded in "Grand Prize Winner list" with a Mac mini with Apple M1 Chip with 8-Core CPU and 8-Core GPU 256GB Storage along with a Certificate of Achievement.

The blogpost will be featured in LearnOpenCV.com.

The page for the LearnOpenCV Blog Olympics 2021 https://lnkd.in/gQVP324Q.