SPRING 2022

Reinforcement Learning on Agroecology (1.0)

Shannon Brownlee *,  Jordan Cahoon *Aryan Gulati,  Leslie Moreno,  Megan Friedenberg,  Colin Ho,  Felix Chen
* Project Lead

Industrial agriculture is unsustainable and destroys local ecosystems. An alternative approach is to instead plant crops within existing microclimates instead of destroying them entirely. This approach has seen success on small scales, such as a Wisconsin farmer successfully growing his crops in a local forest. However, microclimates are very different from one another, and that Wisconsin farmer would have no idea how to succeed in growing crops in a Californian forest, for example. There is no universal method for doing things. CAIS++ students collaborated with Professor Raghavan’s lab to construct a reinforcement learning system using a blackbox environment that can simulate any arbitrary microclimate. Our mission is to model any given microclimate and find the optimal policies for growing crops in that microclimate.

Sepsis Detection

Surya Nehra *Lorena Yan,  Leo Zhuang,  Irika Katiyar,  Sana Jayaswal,  Jaiv Doshi,  Hilari Fan
* Project Lead

The bacteria Staphylococcus aureus is a leading cause of sepsis, a condition which kills 11 million people each year, and contributes to the highest sepsis-related mortality. Most S. aureus strains produce varying amounts of something called alpha toxins (Hla) that bind to and lyse (i.e. rupture) host cells. Studies have shown that the more alpha toxins a S. aureus strain produces, the more deadly it is. Traditional methods to measure an arbitrary strain’s ability to produce toxins by examining lysis of host cells take too long to be useful in a diagnostic setting. In order to speed things up, we will create a machine learning algorithm trained on images of bacteria on blood agar plates. These plates will have varying amounts of cell lysis corresponding to varying measurements of Hla, which is linked to outcomes of infected patients. Such a tool would be much faster than existing methods and could be used in clinical settings.

Targeting Ocean Pollution with Reinforcement Learning (2.0)

Sam Sommerer *Allen Chang,  Anthony Martino,  Priscilla Lee,  Ishu Agrawal
* Project Lead

The Ocean Pollution Project is a collaboration between CAIS++ and Professor Orhun Aydin from the Spatial Science Institute. Using deep reinforcement learning, as well as Lagrangian simulation and optimization, we are analyzing and making decisions about pollution in the ocean. This project was funded by a Microsoft grant.

FALL 2021

ProjectX: How Weather Patterns Influence Disease Outbreaks

Shannon Brownlee *Jordan Cahoon,  Surya Nehra,  Shantanu Jhaveri
* Project Lead

A CAIS++ team represented USC at ProjectX, a machine learning research competition hosted by the University of Toronto. This competition was focused on tackling climate change problems with ML, and our group developed seasonal forecast models that will inform public health systems how future weather patterns will influence disease outbreaks.

Targeting Ocean Pollution with Reinforcement Learning (1.0)

Sam Sommerer *Anthony Martino,  Priscilla Lee,  Ishu Agrawal
* Project Lead

The Ocean Pollution Project is a collaboration between CAIS++ and Professor Orhun Aydin from the Spatial Science Institute. Using deep reinforcement learning, as well as Lagrangian simulation and optimization, we are analyzing and making decisions about pollution in the ocean. This project was funded by a Microsoft grant.