Accelerating Innovation with AI and Machine Learning

When Eric Yang first looked into Northwestern Engineering's Master of Science in Artificial Intelligence (MSAI) program, it was the wide range of topics within the curriculum that most appealed to him. Being able to understand complex concepts like knowledge representation and neural networks intrigued him, and he figured the varied skill set he would develop would be a valuable commodity.

Three years after graduating, Yang confirmed he was right. 

Yang (MSAI '19) is a digital innovator for Sumitovant Biopharma, where he routinely juggles multiple projects at the same time as the company works to accelerate the development of innovative medicines.

"The breadth of techniques and concepts I was exposed to in MSAI allowed me to have a good grasp on high-level techniques and capabilities that exist within machine learning and data science," Yang said. "Though I didn't get deep experience in each, knowing they existed gave me enough direction for where to start researching for a given topic at hand."

Machine learning and artificial learning require continued training and education, so Yang is used to needing to put in extra work to understand a certain concept or framework. At Sumitovant, he frequently finds himself learning more about the healthcare and pharmaceutical space, as well as new technical skills that allow him to continue to grow and evolve in his role. 

At Sumitovant, Yang is embedded within Myovant, one of the company's pharmaceutical subsidiaries focused on addressing unmet needs in women's health. It's his responsibility to work with stakeholders to identify opportunities to build digital solutions to solve their problems.

"My job always keeps me on my toes, so I'm never bored," he said. "I wear a lot of hats and find myself at the intersection of technical internal consulting, data science and software development. Projects can range from ad-hoc analyses using real world data, to developing and maintaining digital applications used by several end users."

Yang has worked on projects that required heavy data engineering, front-end building, and statistical analyses.

"Not all of these were touched on in my formal training," he said, "so being able to learn new skills, languages, and platforms is vital."

Many of the lessons Yang learned during his time in MSAI would sound complex without knowledge and training in machine learning, but one of the most pivotal concepts he took away from the program was also one of the simplest: Garbage in equals garbage out. If the data being used by an application is messy, the application's output will also be messy

"A lot of time and effort needs to be dedicated to cleaning and preprocessing data that is put through training," Yang said. "Instead of spending another hour hyper-parameter tuning, spend that hour further cleaning the data."

Yang frequently reminds himself of that lesson, and he is happy to pass it on to other up-and-coming innovators looking to find their way in the artificial intelligence and machine learning spaces. His other advice? Go to MSAI and leverage all the program has to offer.

"AI doesn't have to be scary," Yang said. "Keep an open mind, and take full advantage of the opportunities in MSAI to work on projects outside of the core classwork."