Research

My research is on learning from noisy labels: training deep networks when some of the labels are wrong. I’m a Ph.D. student in Computer Science at Amirkabir University of Technology, supervised by Dr. Mostafa Shamsi.

• Google Scholar
• Thesis code

Conference papers

• Preventing Overfitting on Noisy Labels Through Adaptive Checkpointing

  A. Azizian, R. Ahmadian, M. Ghatee, M. Shamsi

  10th International Conference on Signal Processing and Intelligent Systems (ICSPIS), pp. 150–154. Shahrood, Iran, 2024.

  • IEEE ICSPIS 2024
  • IEEE Xplore
  • DOI

• Enhanced Multi-Modal Gas Leakage Detection with NSMOTE: A Novel Over-sampling Approach

  A. Azizian, B. Yousefimehr, M. Ghatee

  8th International Conference on Smart Cities, Internet of Things and Applications (SCIoT), pp. 94–99. Mashhad, Iran, 2024.

  • IEEE SCIoT 2024
  • IEEE Xplore
  • DOI

Preprints

• Label Noise Detection and Correction via Ensemble of Siamese Networks

  Master’s thesis, Amirkabir University of Technology. Preprint on OpenReview.

  • M.Sc. thesis
  • OpenReview
  • Code