Hamid Eghbal-zadehView profile
Researcher
Hamid Eghbal-zadeh is a Postdoctoral Researcher at the Institute of Computational Perception at Johannes Kepler University Linz (JKU). Previously, he worked as a Research Scientist at Meta and held roles at the Institute for Machine Learning at JKU. His research focuses on adversarial machine learning, representation learning, and robust deep learning models. He has made significant contributions to acoustic scene analysis, generative models, and reinforcement learning. Education : PhD in Machine Learning from JKU (2019), defended with distinction. BSc/MSc details not explicitly stated in text. Research Highlights : Developed Mixture Density GANs to address mode collapse in GANs Pioneered receptive-field regularization techniques for audio classification Contributed to top-performing systems in DCASE challenges (2016-2019) Active in reinforcement learning and domain adaptation research Awards/Competitions : Multiple top placements in international challenges (DCASE, MediaEval), outstanding reviewer recognitions (ICML 2022, ICLR 2022). Advising : Supervised 6+ students in topics like acoustic scene classification, generative retrieval, and sound event detection.
