Machine Learning Research Scientist
arash.behboodi (at) ti.rwth-aachen.de
Learning Theory, Machine Learning and Deep Learning
Compressed Sensing and Mathematical Signal Processing
Information Theory and Network Information Theory
Since August 2019, I am research scientist (senior staff) at Qualcomm AI research.
Best Technical Paper Award for the paper "A Compressed Sampling for Spherical Near-Field Measurements" at the 40th Annual Meeting and Symposium of the Antenna Measurement Techniques Association (AMTA 2018).
The Best student paper award for the paper "On Generation of Adversarial Examples Using Convex Programming", co-authored by Emilio Balda, Arash Behboodi and Rudolf Mathar - the Asilomar Conference on Signals, Systems and Computers, 2018.
Lecture on Fundamentals of Big Data Analytics.
Milad Alizadeh, Arash Behboodi, Mart van Baalen, Christos Louizos, Tijmen Blankevoort, Max Welling, Gradient l1-Regularization for Quantization Robustness, International Conference on Learning Representations, ICLR 2020
Emilio Rafael Balda, Niklas Koep, Arash Behboodi, Rudolf Mathar, Adversarial Risk Bounds for Neural Networks through Sparsity based Compression, AISTAT 2020, Arxiv