Job Description
This internship will examine Transformer-based Deep Learning Classification Methods for semiconductor defect dispositioning. In the literature in recent years, Visual Image Transformer-based classification networks have been proposed to provide additional model capabilities beyond what traditional Convolutional Neural Networks can provide. In addition to the possibility of increased accuracy, Transformer-based networks can execute object detection tasks and potentially do so with self-supervised training.
A deliverable from the internship will be to select potential Transformer Networks from the literature and evaluate them alongside the traditional approaches our team employs. The intern can use data sets from established projects. Conditional Classification Models are another area of potential benefit for our defect classification work.
Traditionally, models require retraining in order to contextualize the model for a given product. By leveraging a conditional classification approach, it may be possible to retrain only those weights of the model relevant to the product topology changes, yet reuse the other common weights for other detection aspects. This would reduce the training time of the model and reduce overall cost of ownership.
Lastly, the internship will include examining explainability tools beyond the traditional GradCam methods and look at enhanced GradCam and deconvolutional methods to better understand network attention.
Candidate must be currently enrolled in an accredited college or university in industrial engineering, electrical engineering, or computer science/engineering, or similar degree program.
Must have successfully completed up to and including year 2 of a PhD program.
Must be familiar with deep learning program environments such as TensorFlow, PyTorch, ONNX, and OpenVINO.
Must be proficient in Python, and visualization packages therein.
Candidates with prior experience in evaluating deep learning methods for industrially relevant problems are preferred.
Benefits:
We offer a total compensation package that ranks among the best in the industry. It consists of competitive pay, stock, bonuses, as well as, benefit programs which include health, retirement, and vacation. Find more information about all of our Amazing Benefits here:
https://intel.wd1.myworkdayjobs.com/External/page/1025c144664a100150b4b1665c750003
Annual Salary Range for jobs which could be performed in the US:
$63,000.00-$166,000.00Salary range dependent on a number of factors including location and experience.
Work Model for this Role
This role will require an on-site presence. * Job posting details (such as work model, location or time type) are subject to change.
Intel is building on its unrivaled history of innovation to power the future of computing and communications. Intel's ambitions? A world that runs on Intel. Intel is doing this by helping its customers find the technologies and solutions that allow them to transform their businesses and gain market advantage. And it all starts with data. Intel is on a journey from a PC-centric company to a data-centric company. Intel's unmatched depth of experience and scope of vision allows it to dream big, to create a world where Intel powers the future of computing and communications—including powerful processors and accelerators that unlock the full potential of data. Intel continuously delivers advances in performance, power, and connectivity across a diversity of data-centric workloads, so its customers can harness the raw power of data. Intel's innovations span architecture, memory, software, and security to help develop transformative products and experiences for its customers. Collectively, Intel's 107,000 employees worldwide are enabling amazing experiences in 5G, artificial intelligence, driverless cars, and much more.
