Artificial intelligence takes to the ice

My wife does not like clutter. When she sees 50 issues of Nature, the British science magazine, piled on my desk, she gets a little peremptory. Out! she said. I decided on the dignity of an orderly retreat. I had been looking for subjects to write about in what I hope will be a post-COVID world. Call it an act of faith, or at least optimism. So, I sorted through each issue, slicing out  interesting articles and pitching the rest.

In the frenzy of COVID results and explanations over the past 20 months, did I miss anything that would interest people of Northwest Connecticut and the Berkshires? Plenty, it turned out.

There has been a lot of success with gene therapy, specifically the treatment of the immune deficiency that gives rise to babies that have defective immune systems. The same for sickle cell anemia and other blood diseases. There are advances in cancer immunotherapy and interesting articles on antibiotic resistance of bacteria. Our understanding of Parkinson’s disease is advancing.

There was a lot, and we will get to some of these topics over the coming year if COVID continues to recede, which is not a given. We will surely return to COVID, but let’s take a break while we wait to see if the decline in COVID cases continues week to week, whether new therapies affect hospitalization and survival rates and whether mandates work.

And as always in science, there are surprises in store. Just from looking at millions of virus genomes, potential drugs, and the power of the innate immune system, new strategies will appear.  See RichardKessin.com for what I have written, what I have got right and the mistakes I have made.

The Nature article that really caught my attention was comparatively frivolous. That is what I was looking for, a little writerly R&R. It was about curling — 42-pound-stones-with-handles-on-ice curling. Curling is an obsession, not a frivolity, in Norfolk, where we live. The Norfolk Curling Club, of which I am a drinking but not yet a playing member, starts laying down ice at the end of September. It is a fine art because the ice must have a certain consistency and surface.

You may have seen curling in the Olympics or elsewhere. It may be the only Olympic sport in which players can tolerate a bit of a beer belly. Curling clubs always have a bar. Curling is elegant in its own way and has its own lore. The stones, for example, are beautifully milled granite from a quarry in Scotland. Curling is slippery and the stones are heavy, but happily, there are adaptions for the aged. The game is slow and classical, even sedate, and its traditions are not to be trifled with. That is why the article in Nature was strange.

A group of people in South Korea has designed a robot that curls. It sends a rotating stone down the ice with great accuracy. It needs a few throws to get used to the ice, but after that, it is game on. In addition to the condition of the ice, it calculates how to deal with other stones that are in the way. There are no sweepers to direct a moving stone, but there seems to be no need. It is a rather sleek and attractive robot. (According to Wikipedia, the word robot comes from the Czech playwright and journalist Karel Čapek [1880-1938] who introduced it in his 1920 hit play, R.U.R., or Rossum’s Universal Robots.)

How does the robot do compared to elite human competition? The robot’s stones were closer to the target (called the House) in two-thirds of the matches. This is a very quantifiable sport. It is early days though. Count me a bit of a skeptic. The team from the University of Seoul has given us a YouTube video: www.youtube.com/watch?v=uj3ur1uW-7Q.  Maybe Curly the robot would work as a practice partner.

I look forward to competitions between artificial intelligence-powered curling robots. Or perhaps mixed matches with humans. I will not even attempt to explain artificial intelligence, although a friend, Blaz Zupan, a computer expert from Slovenia, once explained it to me. At the time I think I understood, but time is cruel to the aging. Perhaps we will get Blaz to Norfolk during curling season.

Rich Kessin is Emeritus Professor of Pathology and Cell Biology at Columbia University’s Medical Center. He has been writing on science for The Lakeville Journal since 2010 and more recently for The Berkshire Edge and Norfolk Now. See more Body Scientific columns at RichardKessin.com.