About
🧑🏻💻 I am a second-year Ph.D. student in Electrical Engineering at New York University, advised by Prof. Anna Choromanska in the Learning Systems Laboratory.
🔬 My research focuses on self-supervised representation learning and generalization under distribution shift, with autonomous driving as the main testbed. I study how visual pretraining affects the robustness of end-to-end driving models when they are deployed in cities they have never seen, and I work on data-efficient adaptation of self-supervised encoders.
🎓 I received my B.Sc. in Electrical Engineering (Control), with a minor in Computer Engineering, from the University of Tehran.
💼 I am looking for research and ML internships for Summer 2027, in autonomous driving, robotics, or representation learning. Feel free to reach out by email.
Outside of research, I enjoy cooking, exploring New York City, and watching movies.
Selected Publications
View All →Zero-Shot Cross-City Generalization in End-to-End Autonomous Driving: Self-Supervised versus Supervised Representations
Fatemeh Naeinian, Ali Hamza, Haoran Zhu, Anna Choromanska
arXiv preprint arXiv:2603.11417
A controlled zero-shot cross-city stress test for end-to-end driving: I-JEPA, DINOv2, and MAE backbones versus supervised ImageNet backbones under strict geographic splits on nuScenes and NAVSIM.
Mapping Human Grasping to 3-Finger Grippers: A Deep Learning Perspective
Fatemeh Naeinian, Elnaz Balazadeh, Mehdi Tale Masouleh
Proceedings of the 32nd International Conference on Electrical Engineering (ICEE)
Predicts grasp points for a three-finger gripper from RGB images and MediaPipe finger keypoints, reaching 88.2% grasp configurational accuracy.
News
Released the TALKtoME dataset, a curated visual dataset supporting speech and language acquisition in autism.
Our paper on zero-shot cross-city generalization in end-to-end autonomous driving was released on arXiv.
Received the NYU School of Engineering Ph.D. Fellowship.
Started my Ph.D. in Electrical Engineering at New York University.
Presented our work on learning-based robotic grasping at ICEE 2024.