2022 Amazon Fellows
The Science Hub advisory group selected 13 Amazon Fellows for 2022 out of 44 nominations. The fellows are PhD students in the departments of computer science, electrical and computer engineering, mechanical and aerospace engineering, and bioengineering in the UCLA Samueli School of Engineering.
The 2022 Amazon Fellows presented their research during the Lightning Talks event held on Feb 23, 2023. To view the Lightning Talks video or their individual presentation decks, click here.
Sanae Amani Geshnigani
Advisor: Lin Yang
Electrical and Computer Engineering
I develop bandit and reinforcement learning (RL) algorithms for safety-critical decision-making, online advertising, multi-agent and lifelong systems. More specifically, I have worked on bandits in both single-agent and distributed multi-agent settings, safe and lifelong RL with function approximation, and multinomial logistic regression bandits with applications in online advertising.
Advisor: Nanyun (Violet) Peng
Advisor: Prof. Kunihiko Taira
Advisor: Aydogan Ozcan
Advisor: Aditya Grover
Advisor: Dennis Hong
Mechanical and Aerospace Engineering
I formulate software architectures which combines traditional control schemes with vision and machine-learning towards making robots safer and easier to use. Recently, I have been working on auto-tuning methods to self-calibrate low-level controllers and high-level planning algorithms for stable and energy efficient motion planning of legged, aerial, and wheeled robots.
Advisor: Liang Gao
Advisor: Baharan Mirzasoleiman
My research primarily focuses on understanding and improving the efficiency and accountability of machine learning. In particular, I am developing scalable methods to improve the in- and out-of-distribution generalization performance of the models, as well as their robustness against spurious correlations and malicious examples.
Advisor: Guy Van den Broeck
I am interested in enabling machine learning systems to efficiently and reliably learn and infer from noisy, structured, and mixed discrete-continuous data. I approach this goal using probabilistic modeling, statistical relational learning, approximate inference, and formal methods.