Project Description: In advanced semiconductor and other micro-device manufacturing, complex anomalies can arise at various stages of the process. Some of these anomalies can be rare, while still having a large adverse effect on process yield and overall quality. Using machine learning, and in particular deep learning to detect such anomalies is an important recent development in the field, showing promising first results. One major limitation of DL approaches to rare anomaly detection is the need for large labeled samples. In many important use cases such annotated datasets are difficult to obtain on the scale needed for successful training and reliable deployment in the real world. For example, molded semiconductor packages integrated in automotive applications can have micro-scale cracks with major safety consequences, yet such defects are rare and are also hard to detect with current automated inspection systems during manufacturing. Furthermore, the space of unknown yet critical anomalies, may not be fully mapped for new devices. One possible solution to the rare anomaly detection problem is using simulation to generate the needed training data. The goal of this project is to build on the 3D simulation work done at ASM PT for the generation of visually convincing training datasets. While using state of the art graphics modeling and rendering tools provides a good starting point, complex imaging conditions under which devices are inspected in the real world make refinement and further realism in the simulations necessary. The intern will explore the use of generative models, and recent image translation techniques, for the creation of high-quality training datasets. Another key goal is to study the needed mechanisms for the easy control and variation of the simulation output, to allow the generation of the anomaly scenarios of interest. The expected internship work includes:
- In-depth literature study focusing on generative models for simulation
- A selection of key relevant methods for evaluation and further development.
- Proposal for a realistic device simulation framework for at least one relevant ASM PT use case
- Use of ASM PT images for model training and the generation of high-quality simulated datasets
- Demonstration of high-precision defect/anomaly detection based on models trained from simulation data
- Final and mid-term reports containing results and analysis.
References Goodfellow, Ian, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. “Generative adversarial nets.” In Advances in neural information processing systems, pp. 2672-2680. 2014.
Isola, Phillip, Jun-Yan Zhu, Tinghui Zhou, and Alexei A. Efros. “Image-to-image translation with conditional adversarial networks.” arXiv preprint (2017).
Shrivastava, Ashish, Tomas Pfister, Oncel Tuzel, Joshua Susskind, Wenda Wang, and Russell Webb. “Learning from Simulated and Unsupervised Images through Adversarial Training.” In CVPR, vol. 2, no. 4, p. 5. 2017.
Company Description: ASM Pacific is a world-leader in the field of semiconductor equipment. At ASM ALSI in Beuningen, the Netherlands, we develop an advanced multi-beam laser dicing tool, but we also have a technology development team (ASM Center of Competency) that supports the global ASM company with high-tech innovations in various domains, like artificial intelligence, advanced motion control, industrial vision, machine learning and data science. In both the laser dicing business unit and the Center of competency ASM Pacific Technology provides challenging internship positions for university level students. Both short-term (3 months) as well as full-lengths (9 months) MSc internships are supported. We currently have several master students working on their internships in the fields of advanced motion control and artificial intelligence and are always looking for new projects + students.
A dynamic high-tech industrial environment in which you learn to apply academic knowledge to real-life problems
Exposure to the larger ASM organization (>16000 person), opportunities to interact/present to colleagues in Hong Kong and Singapore.
Solid supervision by senior professionals that have a strong academic and industrial background
A market-conform internship compensation and travel expense compensation
Contribute to state-of-the art industrial developments