Varshini Reddy

Propose the use of a mask-conditional synthetic image generation model for creating synthetic satellite imagery datasets and testing on downstream tasks.

Comparison of uncertainty-aware deep learning models on a set of edge-case tasks: robustness to adversarial attacks as well as out-of-distribution and adversarial detection. We find that the geometry of the data sub-manifold is an important factor in determining the success of various models. Our finding suggests an interesting direction in the study of uncertainty-aware deep learning models.

Community detection is applied to system call graphs and features are then extracted from the communities using social network properties such as the centrality measures (degree, betweenness, eigenvector, etc.).

This work involves understanding the importance of the role played by social ties in understanding the causal behavior of customers, and incorporates a novel feature of the social aspects of customers' social group along with the traditional individual customer profiles with potential practical implications, especially in the case of data constraints.

Simulation of cars that can drive autonomously in a diverse set of virtual environments using Convolutional Neural Networks. It is based on lane detection & mapping of steering angle with the surrounding environment.