Yi Ren

Adversarial Attacks on Autonomous Driving with Physically Realizable Patterns

Studying the behaviors of autonomous driving systems when encountering adversarial physical objects will help improve robustness and security.

Program: MORE

Neural ODE Modeling for Many-Body Interactions

Using neural ordinary differential equations to predict stochastic particle behavior will allow for prediction of the start and end behavior of cancer cells and tumors.

Program: FURI

Socially Attentive Autonomous Vehicles

Efficiently modeling high traffic areas will maintain the level of safety in autonomous vehicles and other drivers while reducing the computational cost.

Program: FURI

Learning Unknown Physics Behind Cellular Dynamics Using Time-Variant Neural Network

Creating a time-variant neural network that can model particle dynamics will demonstrate machine learning can learn unknown physics concepts.

Program: FURI

Machine Learning and Dimensionality Reduction to Accelerate DFT Simulations of Material Properties and Enable Computational Material Design

Using machine learning for material property simulations will accelerate the process and decrease computational expense.

Program: FURI

Blame All vs. Blame Me in AVs

Studying the ability for an autonomous vehicle to determine another vehicle's intent or loss of function will promote safer executions of traffic scenarios.

Program: FURI

Adversarial Attacks in Reinforcement Learning for Autonomous Vehicle Control

By attempting to exploit learning systems in autonomous vehicles, we can better characterize the safety of the next disruptive technology.

Program: FURI

The Reinforcement Learning Trojan Horse: Data Poisoning in Autonomous Driving Simulations

Validating the reliability of autonomous vehicle control models will ensure autonomous car machine learning systems can’t be poisoned.

Program: FURI

Crowd-Sourced Service Utilizing Device Network to Facilitate Emergency Intervention

Program: FURI

Energy-Free Personal Security Wristband

Program: FURI