04 February 2019

An article in the Wall Street Journal last Saturday discussed Google's AI system called ARDA (Automated Retinal Disease Assessment) for diagnosing diabetic retinothapy. This sounds like an awesome tool -- and what I love to see -- showing that machine learning will make people's lives better in so many cases. Though, as many machine learning algorithms can be, it's sensitive to data quality. From the article:

"While ARDA is effective working with sample data, according to three studies including one published in the Journal of the American Medical Association, a recent visit to a hospital in India where it is being tested showed it can struggle with images taken in field clinics. Often they are of such poor quality that the Google tool stops short of producing a diagnosis—an obstacle that ARDA researchers are trying to overcome.

"The stakes are high. If diabetic retinopathy is caught early it can be kept at bay through monitoring and management of the diabetes, said R. Kim, an Indian ophthalmologist who runs the Aravind Eye Hospital in Madurai, Tamil Nadu, where Google is testing ARDA. More advanced stages need laser surgery that can stop progression. If it isn’t treated, the condition can cause blindness."

and later in the article:

"If Google allowed the algorithm to make a diagnosis from blurred images, it could miss small lesions that appear in the early stages of the condition, she said. Google must decide how bad an image can be before ARDA refuses to grade it. “It’s a trade-off. We want them to be able to use cameras that are a little harder to use but at some point it should move into something where it is ungradable,” Dr. Peng said."

This seems like an ideal setting for active learning, where the algorithm could request input from doctors when analyzing certain images. The algorithm should also train on blurrier or lower-quality but doctor-labeled images, so that it can learn some of the higher-level features that are indicative of retinopathy.

Still, props to all the companies working hard to improve the health of people around the world. It's a long road and I'm excited people are starting down it.

09 January 2019

Dogs can't operate MRI scanners...

But catscan!! I need to start posting more -- what better way than to start with a bunch of puns!

14 February 2018

like a girl

It's been a long time since I posted! Guess what? I got married! and had a baby! These last few months with my baby girl I have been reminded about the "like a girl" ad from a few years ago. “Yes I kick like a girl, and I swim like a girl, and I walk like a girl, and I wake up in the morning like a girl... because I am a girl.” I can only hope my baby learns how to derive manifold optimization algorithms like a girl.

09 November 2016

for today

“You will never reach your destination if you stop and throw stones at every dog that barks.” — Winston Churchill

10 March 2016

Google's AlphaGo beats a Go champion

For awhile now artificial intelligence has been really good at generating winning game play based on the rules of games. Tic tac toe, checkers, sudoku, chess; in just a few short weeks, one of my students in DSP made an AI that was very good at connect-4. (It didn't beat me-- I was only willing to play it once, though!) However, checking the outcome of every possible move will never be possible for a game like Go unless a completely new paradigm of computing comes about. Still, Google has designed a player that can win. Instead of checking the outcome of moves, AlphaGo learns how best to move based on historical data of real games and its own simulations of games. It keeps trying out new games and it sees what happens. Then it keeps a memory of what to play based on the board configuration using deep learning networks, a sort of dimensionality reduction technique that encodes this experience in several layers of equations that can take any input and give the output of what to play next.

This discussion of what AlphaGo means to the future of AI takes several perspectives on what the implications of a Go-winning AI really are. I agree most with Professor Brunskill. She says, "Go is a fixed game: The rules, possible moves and observable information about the game are all prespecified. AlphaGo is not allowed to invent a new move, nor gain new insight by quizzing its opponent. Fortunately the real world is not like this. From the Hubble telescope to vaccinations, people constantly invent new ideas that allow us to transform how we monitor and shape the universe and achieve previously unimaginable outcomes."

I would add that furthermore, not only because humankind can innovate and create new realities do the rules of life change out from under us. New challenges face us every day, like the disappearance of the Malaysian Airlines flight MH370 two years ago this month, or the Zika virus, or locked cell phones of terrorists. Humans have evolved over centuries to face this adversity head-on and adapt for survival; this is exactly why we do innovate, create new measurement technologies, new drugs, and new security protections. Hopefully computers will be able to help us with this process in the not-too-distant future. But before that, machine learning research needs to face the hurdle of learning in a dynamic world.

13 November 2015

Chess game with the Devil

I had been meaning to read this article about Terry Tao for the last several months and I finally got around to it on a cozy Friday night at home. I really like it for the way it describes him as such a genial and friendly guy, and the story it tells about even Terrence Tao being intimidated when he arrived at Princeton. My favorite quote though is this one about what it's like to be a mathematician:

"The true work of the mathematician is not experienced until the later parts of graduate school, when the student is challenged to create knowledge in the form of a novel proof. It is common to fill page after page with an attempt, the seasons turning, only to arrive precisely where you began, empty-handed — or to realize that a subtle flaw of logic doomed the whole enterprise from its outset. The steady state of mathematical research is to be completely stuck. It is a process that Charles Fefferman of Princeton, himself a onetime math prodigy turned Fields medalist, likens to ‘playing chess with the devil.’ The rules of the devil’s game are special, though: The devil is vastly superior at chess, but, Fefferman explained, you may take back as many moves as you like, and the devil may not. You play a first game, and, of course, ‘he crushes you.’ So you take back moves and try something different, and he crushes you again, ‘in much the same way.’ If you are sufficiently wily, you will eventually discover a move that forces the devil to shift strategy; you still lose, but — aha! — you have your first clue."

20 October 2015

Einstein memorial

Last weekend I was in DC and I visited the Einstein memorial along with all the other national memorials on or near the national mall. There are so many great quotes memorialized on these walls, but this is one of my favorites:

"The right to search for truth implies also a duty: one must not conceal any part of what one has recognized to be true.”