Seyedhamidreza Mousavi

PhD student @Mälardalen University.

prof_pic.jpeg

I am currently a PhD student at Mälardalen University. My research has focused on the automatic design of high-performance, compact, robust, and reliable deep learning-based systems for autonomous driving.

My research interests focus on trustworthy deep learning, encompassing areas such as adversarial robustness, fairness, and reliability in both discriminative and generative models. I am also deeply engaged in tiny deep learning, exploring techniques like pruning, quantization, and neural architecture search to enable efficient deployment of deep neural networks on resource-constrained devices.

news

Jul 30, 2025 My presentation for the Diffusion model in Generative AI Seminar at MDU is available in the blog. Diffusion model
Jun 10, 2025 Our paper got accepted to IJCNN 2025.
Jun 04, 2025 Our presentation for the Reliability of Deep Neural Networks in TSS. Reliability of DNNs

selected publications

  1. AADAM.png
    Aadam: A fast, accurate, and versatile aging-aware cell library delay model using feed-forward neural network
    Seyed Milad Ebrahimipour, Behnam Ghavami, Hamid Mousavi, and 3 more authors
    In Proceedings of the 39th International Conference on Computer-Aided Design, 2020
  2. TAS.png
    Tas: ternarized neural architecture search for resource-constrained edge devices
    Mohammad Loni, Hamid Mousavi, Mohammad Riazati, and 2 more authors
    In 2022 Design, Automation & Test in Europe Conference & Exhibition (DATE), 2022
  3. DASS.png
    DASS: differentiable architecture search for sparse neural networks
    Hamid Mousavi, Mohammad Loni, Mina Alibeigi, and 1 more author
    ACM Transactions on Embedded Computing Systems, 2023
  4. PROACT.png
    ProAct: Progressive Training for Hybrid Clipped Activation Function to Enhance Resilience of DNNs
    Seyedhamidreza Mousavi, Mohammad Hasan Ahmadilivani, Jaan Raik, and 2 more authors
    arXiv preprint arXiv:2406.06313, 2024
  5. proard.png
    ProARD: progressive adversarial robustness distillation: provide wide range of robust students
    Seyedhamidreza Mousavi, Seyedali Mousavi, and Masoud Daneshtalab
    arXiv preprint arXiv:2506.07666, 2025