PhD - Leuven | More than two weeks ago
Make machine learning into a helpful tool to gain insight in electronic device reliability phenomena.
In recent years, Artificial Intelligence (AI) has made impressive progress and many applications are entering the daily lives of everyone. Expectations are that this is merely the beginning of a whole new era in which exciting new applications will arise in fields as for instance medicine, self-driving cars, drug design, robotics, arts, etc...
Also in a research environment, machine learning (ML) techniques are being deployed as versatile tools to extract more information from experimental data and from simulation models. In this PhD, the student explores the opportunities offered by machine learning techniques for semiconductor device reliability research.
The student takes a top-down approach: starting from existing machine learning algorithms, he/she looks for opportunities to apply these to enhance the insight in or facilitate the study of degradation physics.
Possible opportunities include (non-exhaustive list):
-use of ML to model simulation results of complex physical situations. This could allow for interpolation in a multidimensional parameter space without having to perform time-consuming additional simulations.
-ML models for classifying and recognizing degradation regimes with complex workloads.
-ML models to understand device-to-device variations, as for example modeling the noise in electrical signals generated by underlying configurations of semiconductor dopant fluctuations or dielectric defects.
-further opportunities as they come along...
It is important that the student understands the conceptual ideas behind machine learning. Given the nature of a research application, the focus in this PhD will be on algorithms for fast learning from few data points. A simple ‘black box’ approach will not be sufficient. Instead, the student will need to adapt or (re)design ML models to obtain efficient learning systems. These adaptations include for example: shaping the structure of the model based on physical structural knowledge and properties, selecting physically meaningful features, finding principal components in data, reducing the number of parameters and dimensions to a minimum in order to avoid overfitting, etc... To this purpose, a creative and innovative attitude is required, and the student should also acquire a thorough insight in the physics of the reliability phenomena that are the topic of the ML model. The student will work at imec in the Advanced Reliability, Robustness and Testing Department, and is embedded in a team of semiconductor device reliability specialists.
At the end of this PhD, the student should come to a realistic picture of the gains and limitations of machine learning techniques in reliability research.
Target students should be motivated and creative in their approach. They can have a background in either electrical engineering, physics, mathematics, or similar.
Required background: electrical engineering, semiconductor physics, mathematics
Type of work: 50% simulation/modeling, 30% literature, 20% experimental
Supervisor: Hendrik Blockeel
Co-supervisor: Robin Degraeve
Daily advisor: Robin Degraeve
The reference code for this position is 2022-006. Mention this reference code on your application form.