[ad_1]
FORT COLLINS, Colo. — Artificial intelligence that helps protect honeybees is more than a science fiction fantasy, and it’s a reality at Colorado State University, funded by the National Science Foundation.
An interdisciplinary team, including partners from CSU Extension, the School of Computer Science, and the School of Education, received an $845,000 grant from NSF to harness the power of community science and new AI models to study more than 900 pollinators. Assisted in monitoring and recording health conditions. This species is found throughout Colorado.
“This kind of research is really new frontiers, and if all goes well, we’re just getting started,” said Nathaniel Blanchard, assistant professor of computer science at CSU.
Bees play an important role in human life, Three-quarters of the crops we eat are pollinated, and their conservation is paramount in the face of a changing climate. As recently announced by Governor Jared Polis in collaboration with CSU, pollinator health is at the forefront of public policy in Colorado. Health survey of Colorado’s native pollinating insects.
Native Bee Watch Scaling
Native Bee Watch (NBW), a community science biodiversity monitoring program, found a permanent home at CSU Extension in 2019. Extension aims to fulfill CSU’s land-grant mission by providing cutting-edge research and education to communities across Colorado.
NBW uses volunteer power to record and monitor honey bee activity across the state, supporting homeowners, ecologists, and municipal planners in their efforts to protect wild honey bee populations. During the pandemic, the program moved online and became available across the state, and the move attracted 300 new volunteers. This is a significant increase from the program’s previous 20 students per year.
“The level of interest from our volunteers proves how important pollinator protection is to Coloradans,” said Lisa, NBW founder and director and CSU Extension horticulture specialist. Mason said. “We couldn’t make this program possible without the dedication of our volunteers. Their excitement is contagious, especially when they start seeing bees that have just learned to identify them.”
Participation growth was so rapid that NBW’s virtual training for new volunteers could not scale fast enough. Thus, a collaboration with CSU’s Department of Computer Science was formed. What’s the mission? Personalized AI algorithms. Train new volunteers using the latest pedagogy and agile AI via the app, tailoring the curriculum to a specific volunteer’s experience level.
Creation of novel AI
To create this new AI, CSU computer scientists must complete three important tasks. First, the AI model must be able to identify bees based on images and input from a wide range of volunteers. This is easier said than done because many bees can only be identified to species level under a microscope and many volunteers are still learning. How to properly identify the bees they are seeing.
“If we had enough AI to identify bees on their own, we could use it to train volunteers and validate their observations, creating research-quality ecological datasets. ,” said Associate Professor Jill Zarestky. The School of Education says, “…then we can begin to talk about the prevalence and distribution of native bees in Colorado.”
Second, the model must support adult learners from diverse backgrounds, estimate their expertise, and provide lessons and feedback tailored to the learner’s current background. This is called explainable AI. The quality of AI feedback is essential to ensure that the data collected by volunteers is scientifically and ecologically accurate.
The challenge for computer scientists lies in the diversity of learners. Unlike a classroom environment, there are no control or known factors regarding volunteers and their level of knowledge. It also lies in the diversity of bee species and the large number of flies and wasps that mimic them. Bees as a defense mechanism.
Even if a novice volunteer sees a wasp or a fly, he or she may realize that he or she has observed a bee and upload a photo of that observation. Adding a photo gives the AI a chance to evaluate the accuracy of the volunteer’s data. AI ensures data integrity by automatically detecting identification errors, and provides additional information, including tips on how to tell bee imitations from the real thing and how to collect high-quality evidence to support bee imitations. Pinpoint specific areas where users can benefit from training. Claim.
The final component is creating a human-AI interface, an app that can provide adaptive training.
Faculty of Education researchers provide expertise on how to best classify and train different types of learners.
“‘What is the right way to structure a lesson?’ ‘What questions should I ask?’ These are all decisions made by education experts,” said Sarath Reedharan, assistant professor of computer science at CSU. “We’re going to build this on very sound educational theory, but how do we translate those educational concepts first into mathematics and then into software? We are focusing on the theory and software of this project, the foundations on which the final tools will later be built.”
Implications from moment to existence
The implications of this research range from the momentary to the existential.
“Community social science gets people outside,” Mason said. “This allows people to slow down and look at their environment,” he said. “One of the things I love about pollinator and insect conservation is that everyone can be part of the solution. ”
From a community perspective, such activities enable two-way communication and input between the community and the university.
“I think this is an opportunity for the research expertise to be reflected in the expansion, then reflected in the community, and then the community to feed back into the research,” Zarestky said. “It’s not a loop that we close often.”
At the national level, the team hopes their algorithm will make community science more viable and scalable.
“We hope this project provides an automation template that can be easily adapted to other community science and adult science education programs,” said Nikhil Krishnaswamy, assistant professor of computer science at CSU.
In addition to expanding this research to support other community science programs, this research has more existential implications regarding human and AI learning.
“AI is often in the news for its association with creating misinformation and spreading propaganda, so the application of AI to areas such as community science will help people think more scientifically and reason better.” I hope that I can show you a different paradigm of being a tool that can help you,”’ Krishnaswamy said.
Native Bee Watch is open to anyone across the state. To stay informed and join the program, visit NativeBeeWatch.org and sign up to receive our newsletter.
— Ally Lachman
colorado state university
[ad_2]
Source link
