Ithaca, N.Y. — Today we present our third installment of an Ithaca Voice series highlighting just a handful of the crazy cool things Cornell University professors are researching, writing, designing, discovering, or (insert here) at any given time.
We have no doctrinal preferences and no academic prejudices. Our sole criteria is that the professor’s work be, as the headline suggests, “crazy cool.” And, no, we don’t have a precise definition of “crazy cool.”
(Got a professor we should highlight? Email firstname.lastname@example.org.)
Without further ado…
1 – Cornell study reveals Southwest U.S. ‘megadrought’ possible within century
A Cornell professor-led study revealed this week that due to global warming, there is at least a 50 percent chance that the southwestern United States will experience a decade-long drought within the next century.
The study, in conjunction with the University of Arizona and U.S. Geological Survey researchers, showed that there is also between a 20-50 percent chance that this “megadrought” could last up to 35 years.
“For the southwestern U.S., I’m not optimistic about avoiding real megadroughts,” said Toby Ault, the lead author of the study and assistant professor at Cornell. “As we add greenhouse gases into the atmosphere – and we haven’t put the brakes on stopping this – we are weighting the dice for megadrought.”
“With ongoing climate change, this is a glimpse of things to come. It’s a preview of our future,” Ault said.
2 – Thai-expert publishes book of essays
Cornell professor Benedict Anderson has published a book of essays that a colleague says establishes “the tone and framework for understanding the American Era (late 1950s to the early 1970s) in Thailand.”
“Exploration and Irony in Studies of Siam Over 40 Years” explores the work of Anderson in the field of Southeast Asian Studies.
From exploring the role of the Thai military to its monarchy to its art, Anderson’s book critiques Thai society and its most powerful institutions.
3 – Surgeries on animals could lead to human repairs
Surgeons at Cornell’s University Hospital for Animals attempted a new method of cartilage repair on horses last week.
Next week, another team will attempt to repair a meniscus on sheep, using a 3D printer to assemble an artificial meniscus that fits the patient’s body.
“The goal is to make these technologies available for people,” said Lisa Fortier, professor of large animal surgery at Cornell’s College of Veterinary Medicine.
“If they do well in these animals, then the [Food and Drug Administraion] can approve it for use in humans,” she said.
4 – “Robo Brain” uses Internet to teach robots
“A giant repository of knowledge,” known as “Robo Brain,” uses information collected from the Internet that is stored in a robot-friendly format to do basic tasks – finding keys, putting away dishes, and pouring drinks, for example.
“Our laptops and cell phones have access to all the information we want,” said Ashutosh Saxena, assistant professor of computer science. “If a robot encounters a situation it hasn’t seen before it can query Robo Brain in the cloud.”
By consulting the Robo Brain, robots with no knowledge of certain tasks can teach themselves how to do those tasks.
In conjunction with professors from Brown, Stanford, and UCLA-Berkeley, Saxena has helped compile one billion images, 120,000 YouTube videos and 100 million how-to documents and appliance manuals – all pieces of information that the robot can use to recognize objects and how they are used, along with human language and behavior.
Read an excerpt from The Cornell Chronicle below:
The system employs what computer scientists call “structured deep learning,” where information is stored in many levels of abstraction. An easy chair is a member of the class of chairs, and going up another level, chairs are furniture. Sitting is something you can do on a chair, but a human can also sit on a stool, a bench or the lawn.
A robot’s computer brain stores what it has learned in a form mathematicians call a Markov model, which can be represented graphically as a set of points connected by lines (formally called nodes and edges). The nodes could represent objects, actions or parts of an image, and each one is assigned a probability – how much you can vary it and still be correct. In searching for knowledge, a robot’s brain makes its own chain and looks for one in the knowledge base that matches within those probability limits.
“The Robo Brain will look like a gigantic, branching graph with abilities for multidimensional queries,” said Aditya Jami, a visiting researcher at Cornell who designed the large-scale database for the brain. It might look something like a chart of relationships between Facebook friends but more on the scale of the Milky Way.