Interns at HP Labs are encouraged to apply theoretical insights, creativity, and their problem solving skills to real world technical and business problems. For He Luan, that has meant spending a summer in HP's Print Adjacencies and 3D Lab exploring how deep learning methodologies can be used to improve 3D printing.

Luan became interested in 3D printing while majoring in robotics and intelligent systems at the University of Science and Technology of China. One class in particular, she recalls, “gave me an idea of how revolutionary 3D printing could become in terms of how we make things and also what the challenges were to improving it.”

The insight spurred Luan to enroll in the Ph.D. program in industrial and systems engineering at the University of Southern California, where she specializes in applying data and statistical learning to 3D printing research.

As an HP labs graduate intern, Luan gets to deploy her theoretical knowledge towards a very concrete end. A major research project at HP Labs, she notes, is investigating ways to analyze sensor data from HP 3D printers to help optimize the printing process.

“As part of that effort, I’m applying deep learning to these data sets and exploring different neutral network architectures to predict the thermal behavior of each layer of material as it is printed,” Luan explains.

It’s challenging work. More conventional deep learning problems, like image identification, can tolerate a certain degree of imprecision in their predications. But to be useful in 3D printing, predictions need to be accurate to an individual voxel (the 3D equivalent of a pixel).

In addition, thermal behavior in a 3D printer is incredibly complex. Other HP Labs teams, however, are conducting multiple physical experiments to uncover these interactions and sharing that data with Luan.  

“By applying deep learning to that data, we’re hoping to reveal information patterns that allow us to predict thermal behavior and help us to build even more accurate physical models,” she says.

Luan hopes that results from her internship will find their way into her dissertation, which explores new ways to improve geometric accuracy in 3D printing. And she’s hoping to continue working with HP Labs after her internship.

“I’d love to help some of the ideas we cook up over the summer find their way into products,” she says.

HP Labs interns typically work over a 10-12 week period between undergraduate or graduate semesters and must be enrolled full-time at an accredited college or university (US: minimum 12 credit hours per semester). Some HP Labs interns are able to continue their research over successive internships and many are subsequently hired as full-time employees within the HP Labs organization.