Eye on AI - December 27th, 2021
Welcome to Aigora's "Eye on AI" series, where we round up exciting news at the intersection of consumer science and artificial intelligence!
This week our focus is AI-led conservation, including how one professor’s obsession with game theory led to a multitude of AI-based solutions that combat environmental threats and a robot that’s scaring invasive fish species into regression.
How One Professor’s Obsession with Game Theory Is Reducing Environmental Threats
When Fei Fang was introduced to game theory (mathematical models that describe strategic interactions among rational decision-makers) as an electrical engineering graduate student at USC, she knew she’d found her calling. Already armed with a bachelor’s degree in Biology, she saw in game theory an opportunity to reduce environmental and terrorist threats, and began testing her theories on computational game theory by helping the U.S. Coast Guard prevent terrorist attacks.
“We had to deal with an infinite number of patrol routes the U.S. Coast Guard patrol boats could take,” says Fang. “The attacker could also strike the ferries at any time. This makes the problem very different [from airport security] and more challenging because there are an infinite number of actions possible on both sides.”
The result was the development of a compact representation in which the AI algorithm considers all the times and locations terrorists might attack, which led to the creation of the PROTECT (Port Resilience Operational/Tactical Enforcement to Combat Terrorism) model for the U.S. Coast Guard in 2013, which is still in use today.
Next, Fang turned to her biological roots by searching for ways her model could reduce poaching in wildlife preserves. This led to the creation of the machine-learning system PAWS (Protection Assistant for Wildlife Security), which uses data from past patrols to predict where poaching is likely to occur.
“Fang and her colleagues refined their system and began testing the improved tool, which took the topography into account when generating recommended patrol routes,” writes IEEE Spectrum contributor Joanna Goodrich. “... This year Microsoft added PAWS to its Azure platform, a portfolio of AI services designed for developers and data scientists. It also was integrated into SMART, a platform that consists of software and analysis tools designed to help conservationists manage and protect wildlife. SMART is used at more than 600 conservation sites around the globe.”
Now an assistant professor at Carnegie Mellon, Fang has turned her attention to food insecurity. She and her research team at Carnegie Mellon’s Institute for Software Research are developing an algorithm to help 412 Food Rescue, a nonprofit that uses a smartphone app to notify volunteers who are within 8 kilometers of the pickup zone when there is food available, to increase the pickup rate of good but unsellable food from grocery stores.
“The new system was deployed in February 2020,” continues Goodrich. “In one month, the pickup rate increased from 84 percent of food being claimed to 88 percent. The time it took for a task to be accepted decreased from 78 minutes to 43 minutes.”
From terrorist and poaching prevention to food insecurity assistance, it’s remarkable what Fang has already accomplished using variations of her game theory model. Let’s hope other researchers take her lead to use game theory in broader AI applications.
Robot Scares Invasive Mosquitofish Into Hiding
Continuing with the theme of conservation, we turn now to the article “Robots use fear to fight invasive fish,” which describes how researchers from Australia, the U.S., and Italy teamed together to develop a robot that mimics the largemouth bass to scare the invasive mosquitofish into regression.
“Mosquitofish is one of the 100 world's worst invasive species, and current methods to eradicate it are too expensive and time-consuming to effectively contrast its spread,” says first author Giovanni Polverino (@GioPolverino) of the University of Western Australia. “This global pest is a serious threat to many aquatic animals. Instead of killing them one by one, we're presenting an approach that can inform better strategies to control this global pest. We made their worst nightmare become real: a robot that [mimics the largemouth bass and] scares the mosquitofish but not the other animals around it."
Using a test environment, the research team found that after only a brief encounter with the robot fish, mosquitofish populations huddled together for protection and were more reluctant to explore the rest of the enclosure. They became less active, ate more, and froze longer, all signs of anxiety that continued weeks after their last encounter with the robot. Conversely, the tadpoles the mosquitofish had preyed upon became more active, exploratory, and healthy, all indications of reduced anxiety. Furthermore, researchers began to see that the robot fish also reduced mosquitofish populations.
“After five weeks of brief encounters between the mosquitofish and the robot, the team found that the fish allocated more energy towards escaping than reproducing,” writes MIT’s editorial team. “Male fish's bodies became thin and streamlined with stronger muscles near the tail, built to cut through the water for fleeing. Male fish also had lower sperm counts while females produced lighter eggs, which are changes that are likely to compromise the species' survival as a whole.”
While the new robot fish isn’t ready to be released into the wild, it’s a promising first step in the fight to combat invasive fish species, which is a huge problem throughout the world, and could soon offer a solution to protect freshwater ecosystems.
That's it for now. If you'd like to receive email updates from Aigora, including weekly video recaps of our blog activity, click on the button below to join our email list. Thanks for stopping by!