Saasha Nair | Research

image
👩‍🔬

Research Interests

With Machine Learning based solutions seeping into our everyday lives, it becomes necessary to ensure that the products we design are safe and reliable. Thus, my interests in AI lie at the intersection of Reinforcement Learning and Safety.

📜

Publications

On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Driving.

Rossolini, G., Nesti, F., D'Amico, G., Nair, S., Biondi, A., Buttazzo, G., Preprint on ArXiv. 2022.

Evaluating the Robustness of Semantic Segmentation for Autonomous Driving against Real-World Adversarial Patch Attacks.

Nesti, F., Rossolini, G., Nair, S., Biondi, A., Buttazzo, G., Winter Conference on Applications of Computer Vision (pp. 2826-2835), 2022

An Evaluation of “Crash Prediction Networks” (CPN) for Autonomous Driving Scenarios in CARLA Simulator. (Won Best Paper Award)

Nair, S., Shafaei, S., Auge, D., & Knoll, A., The AAAI's Workshop on Artificial Intelligence Safety, 2021

Falsification-based Robust Adversarial Reinforcement Learning.

Wang, X., Nair, S.,& Althoff, M., IEEE International Conference on Machine Learning and Applications, 2020

Monitoring Safety of Autonomous Vehicles with Crash Prediction Networks.

Nair, S., Shafaei, S., Kugele, S., Osman, H., & Knoll, A., The AAAI's Workshop on Artificial Intelligence Safety, 2019

A Machine Learning Based Approach to Application Landscape Documentation.

Landthaler, J., Uludağ, Ö., Bondel, G., Elnaggar, A., Nair, S., & Matthes, F., IFIP Working Conference on The Practice of Enterprise Modeling(pp. 71-85), 2018

image