Goals
- Creating and improving classifiers which will predict cichlid fish behavior in video data.
Biology: Review the literature to find behaviors of biological importance. Learn to identify complex behaviors in video which aids in correctly training our machine learning model.
Computer Science: Learn and utilize computer vision classifier. Seek and develop new tools that will improve accuracy on cichlid behavior.
Math: understand classification algorithms and provide a comparative analysis of why one algorithm performs better than another for our application. - Use unsupervised clustering techniques to extract biologically significant expression of behavior using neuronal data.
Biology: Understand the significance of neuronal cell types, genes, and traditional methods for studying gene expression
Computer science: complete a robust comparative study of clustering techniques and work with biologist to determine the method with highest biological insight.
Math: understand clustering algorithms and provide a comparative analysis of why one algorithm performs better than another for our application. - Create a fish simulation that cichlids can interact with in aggression studies.
- Generate a breeding simulation for producing cichlids hybrid with evolutionary significant traits.
- Create subject-conscious education materials for biology students learning mathematics and computer science.
Issues Involved or Addressed
Use machine learning to robustly understand the connection between observed behavior and brain function. Students will be working in groups arranged according to field. Each project will have a biological, mathematical, and computational component. Student will rely on the work and expertise of other in their group to communicate how the biology, math, and computer science should be considered with respect to their problem.
Methods and Technologies
Academic Majors of Interest
- Computing›Algorithms, Combinatorics and Optimization
- Computing›Computer Science
- Computing›OMSCS synchronous
- Physics
- Sciences›Biology
- Sciences›Mathematics
- Sciences›Neuroscience
- Sciences›Statistics
Preferred Interests and Preparation
The project would be great for students from either a biology, computer science, or math background. We do not expect an expertise in all disciplines. These are highly collaborative projects. Math Students: should have strength in statistics and linear algebra and a passion for applied research. You may work in Matlab if you'd like computer science students: Project will be completed in Python and R One project has a website development component. Biology: Interest in the following topics: neuroscience, genetics, ecology , and evolution bonus if they know python or R
Meeting Schedule & Location
Team Advisors
- Biological Sciences
- Biological Sciences