Daphne Koller leads Insitro, a company applying machine learning to pharmaceutical drug development. She is also a co-founder of Coursera (one of the world’s largest online education platforms with 30 million + users), a MacArthur Fellowship recipient, and Computer Science Professor at Stanford University.
Recently, she discussed machine learning in the biomedical industry as well as what she thinks are vital characteristics to achieve individual success with Iridescent CEO & Founder Tara Chklovski.
Tara Chklovski (TC): Can you share two of the most interesting problems that you’ve tried to tackle in your career and what you’re working on now?
Daphne Koller (DK): My work cuts across a variety of fields including biomedical, education, and problems like computer vision. There are two specific biomedical projects I’m very proud of. The first one involved working with a neonatologist, a doctor for premature babies, to predict survival rates for neonates. Neonates are premature, teeny babies – they’re about the size of your hand and weigh less than 200 grams. They often struggle to survive. The doctor and I worked together to predict the risk of death for these babies. Our hope was to help doctors identify and help the babies at highest risk survive. In our study we didn’t rely on the traditional tests for babies’ health (which we found don’t work well for babies born that early), and we didn’t want to rely on invasive measures like sticking these tiny babies with needles! So we used the data from the bedside monitor. The bedside monitor measures things like the baby’s heart rate, respiratory rate, and their oxygen saturation. It turns out that there are markers in that non-invasive data that can predict survival rates, and that you don’t need that much data to make useful recommendations. We were able to make predictions based on data collected from the first few days of the baby’s life.
Another project I did with a PhD student of mine who was an MD/PhD pathologist. We were looking at images of tumors from breast cancer patients and trying to predict five-year survival rates to help organize patients by risk. We took a data-driven approach. Pathologists have been looking at these sort of images for a century. For our project, we included more features in our dataset than the standard features pathologists consider. We put in everything that we could think of, including hundreds of features that no one has ever looked at before. As a result, our predictions were better than pathologists’, and the features that were being used by our predictor were quite different from the ones that pathologists had been looking at. Specifically, they had been looking at features of the tumor cells, and it turned out that the features of the cells that surround the tumor were actually more predictive! Today, this is well recognized as a critical factor in cancer survival, and is called the tumor microenvironment. This microenvironment where cancer lives, and the immune system are factors that are absolutely critical in cancer patient survival. Our paper was one of the earliest pieces of evidence supporting that.
TC: Can you share more about your latest venture?
DK: Yes, I started a company called Insitro to address the crisis of increasing drug development costs. The cost of developing a new drug is now estimated at $2.3 billion. This is a problem and a key component in the increasing cost of healthcare. We are exploring ways to use machine learning to address this problem.
One of the things that excites us about this project is the data-rich ways that machine learning can make a big difference. Biology has historically been a data-poor industry so the application of machine learning was limited. But now, there’s an incredible increase in the ability of biologists to produce very large amounts of robust data in a scalable way as a result of a combination of technologies like automation, miniaturization, microfluidics, CRISPR, microscopy, and organoids.This makes machine learning a more viable tool in this field.
Now, we’re able to leverage all these technologies and combine them in innovative ways for experiments and produce data for machine learning. That’s never been possible before. I think that is a huge opportunity for machine learning to make a difference in the field.
TC: That’s exciting to hear, how we’re now able to solve problems in biology that were previously inaccessible. In our AI Family Challenge program we encourage families to identify community problems to solve. What advice would you give to children and adults trying to find a good problem to tackle?
DK: I would encourage them to think about problems that they personally care about and believe need a solution. Problems that make them stop and say, “Wouldn’t it be great if we didn’t have to deal with this annoying problem every day?” or “Wouldn’t it be great if we could automate this thing that’s been very time-consuming for us to do”? The second thing they should look for is an area where there is enough data for a machine learning model to be built.
TC: Looking forward, how do you think AI can help strengthen societies?
DK: I think AI is a fundamental technology that enables us to use computers to solve problems that are so hard that we don’t know how to solve ourselves. That includes everything – from recognizing images, to designing cars that drive themselves, to designing better drugs. People have been bashing their heads against those problems for a very long time and not succeeding.
There are so many problems around us that are really important. People should go and seek them out. For instance, I know people who are using machine learning to design drones. That has benefits both from a fairly mundane ecommerce perspective, like getting your groceries or your shopping delivered to your home, but also from a broader reach, like delivering drugs to remote places. I think there’s a lot of places where there are significant societal problems that can be addressed, using a powerful technology like AI.
TC: How do you personally overcome challenges or barriers when you encounter them in your work?
DK: I think it’s important to keep in mind that what you’re doing is (hopefully) important to you and important to society. Every important thing has its ups and downs, and you’ll need to work through them and persevere. A lot of people give up too quickly when challenges come up, and — inevitably, challenges do arise. Usually the really successful people are so, not because they’re more talented, or even because they have a better idea, but simply because they stick with it. Grit, as Angela Duckworth puts it, is very important. It’s very important in school, it’s important in one’s career, and it’s important in one’s life. You have to be willing to take risks. A lot of people find themselves wishing they had done something that was more adventurous, something that was more aspirational, but they were too scared to leave their comfort zone and do something that was different, something that didn’t have as much of a safety net. I know very few people who tried something bold and aspirational and regret it, even if it failed.
I’ve seen that in my own career.The early days of Coursera were very challenging. I mean, we were all new to the business, we were doing something that was very different, and in some cases, very unpopular. Not every university thought it was a good idea to put content up online for free. There were a lot of challenges, a lot of people who objected, and moments along the way where it seemed like things were going to crumble. It would have been very easy for us to give up and go back to cushy faculty positions. But we stuck through it, and if we hadn’t, Coursera wouldn’t exist, and 35 million learners would not have been helped.
TC:A lot of the parents we work with aren’t very familiar with technology. How would you motivate and help them to see that they can positively change the world with limited technology experience?
DK: They need to get started on small things that make them understand their self-worth, and as you build up that sense of self-worth, it becomes increasingly easier. It’s kind of like building a muscle.
And just don’t give up!