Machine Learning

The Computational Sensing team at MERL is seeking motivated and qualified individuals to develop algorithms that can perform background oriented schlieren (BOS) tomography. The project goal is to utilize both analytical and learning-based architectures to enable the reconstruction of 3D air flows in an indoor setting from BOS measurements coupled with physics informed machine learning. Ideal candidates should be Ph.D. students and have solid background and publication record in any of the following, or related areas: imaging inverse problems, large-scale optimization, differentiable scene rendering, learning-based modeling for imaging, and physics informed neural networks. Preferred skills include experience with schlieren tomography, inverse rendering, neural scene representation, and computational imaging hardware. Publication of the results produced during our internships is expected. The duration of the internships is anticipated to be 3-6 months. Start date is flexible.

Required Specific Experience

• Experience with differentiable/physics-based rendering.

MERL seeks graduate students passionate about deep learning and energy systems to contribute to the development of deep time-series forecasting models for real building energy data. The work will involve multi-domain research including deep learning model development, time-series analysis, and possibly integration with energy management systems. The methods will be implemented and evaluated using real-world datasets. The results of the internship are expected to be published in top-tier machine learning and energy systems conferences and/or journals.

Exact start date is flexible (most likely Summer 2025), with an expected duration of 3-6 months, depending on agreed scope and intermediate progress.

Required Specific Experience:

• Current or past enrollment in a PhD program in Electrical Engineering, Computer Science, or a related field with a focus on Machine Learning or Energy Systems.
• 2+ years of research experience in at least some of the following areas: deep learning, time-series analysis, probabilistic machine learning, energy systems modeling.
• PyTorch fluency.
• Familiarity with real-world data wrangling.
• Experience with time-series data visualization and analysis tools.

Strong Pluses:

• Familiarity with transformer-based time-series forecasting methodologies e.g. TFT or time-series foundation models.
• Familiarity with adaptation mechanisms e.g. fine-tuning, meta-learning.