About Me
Greetings! I'm Sajjad Amini, proudly serving as an Assistant Professor at the Electronics Research Institute, Sharif University of Technology. As a dedicated professional, I specialize in Machine Learning, with a keen focus on Deep Generative Models and Robust Deep Learning, constantly striving to advance knowledge and innovation in these dynamic fields. At Sharif University, my work revolves around groundbreaking research in adversarial attacks and defenses, deepfake detection, and advancing contrastive learning in speech processing. My teaching portfolio features courses like Introduction to Machine Learning and Linear Algebra. Alongside my academic pursuits, I have authored several papers in esteemed journals and conferences, emphasizing the robustness of deep neural networks and other innovative topics.
Publications
- Journal Papers
- Amini, S., Heshmati, A., & Ghaemmaghami, S. (2024). Fast adversarial attacks to deep neural networks through gradual sparsification. Engineering Applications of Artificial Intelligence, 127, 107360.
- Heshmati, A., Amini, S., Ghaemmaghami, S., & Marvasti, F. (2022). Designing Low Coherent Measurement Matrix With Controlled Spectral Norm Via an Efficient Approximation of ℓ∞ -Norm. IEEE Signal Processing Letters, 29, 2243-2247.
- Amini, S., Soltanian, M., Sadeghi, M., & Ghaemmaghami, S. (2022). Non-Smooth Regularization: Improvement to Learning Framework Through Extrapolation. IEEE Transactions on Signal Processing, 70, 1213-1223.
- Amini, S., & Ghaemmaghami, S. (2020). Towards improving robustness of deep neural networks to adversarial perturbations. IEEE Transactions on Multimedia, 22(7), 1889-1903.
- Soltanian, M., Amini, S., & Ghaemmaghami, S. (2019). Spatio-temporal VLAD encoding of visual events using temporal ordering of the mid-level deep semantics. IEEE Transactions on Multimedia, 22(7), 1769-1784.
- Amini, S., & Ghaemmaghami, S. (2019). Lowering mutual coherence between receptive fields in convolutional neural networks. Electronics Letters, 55(6), 325-327.
- Amini, S., & Ghaemmaghami, S. (2019). A new framework to train autoencoders through non-smooth regularization. IEEE Transactions on Signal Processing, 67(7), 1860-1874.
- International Conferences
- Amini, S., & Ghaemmaghami, S. (2022, May). Towards Robust Visual Transformer Networks via K-Sparse Attention. In ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 4053-4057). IEEE.
- Amini, S., & Ghaernmaghami, S. (2018, September). Sparse autoencoders using non-smooth regularization. In 2018 26th European Signal Processing Conference (EUSIPCO) (pp. 2000-2004). IEEE.
- Amini, S., Sadeghi, M., Joneidi, M., Babaie-Zadeh, M., & Jutten, C. (2014, September). Outlier-aware dictionary learning for sparse representation. In 2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP) (pp. 1-6). IEEE.
