Dr. Seyedeh Tina Sefati | Reinforcement Learning | Best Researcher Award
Dr. Seyedeh Tina Sefati | Reinforcement Learning | University of Tabriz | Iran
Dr. Seyedeh Tina Sefati is a highly skilled and innovative Ph.D. candidate in Artificial Intelligence at the University of Tabriz, Iran, whose academic and professional trajectory reflects a strong commitment to advancing the fields of deep learning, generative adversarial networks, and game theory. Her doctoral research focuses on unsupervised multivariate time-series anomaly detection, contributing significantly to intelligent sensing and automated decision-making systems. Dr. Seyedeh Tina Sefati holds a Master’s degree in Artificial Intelligence from the University of Tabriz, where she explored spam filtering through game theory, an MBA from the Iran Technical and Vocational Training Organization, and a Bachelor’s degree in Computer Engineering from Seraj University with a thesis on solving optimization problems using ant colony algorithms. Professionally, Dr. Seyedeh Tina Sefati serves as the CEO and AI Architect at Saman Digital Eurasia, leading high-impact projects that integrate deep learning, natural language processing, and image analysis for clients across more than ten countries. Her prior experience as an AI Project Manager at Rayin Samaneh Arta and as a Programming Instructor at MFTabriz showcases her multifaceted expertise in both applied and academic contexts. Her research interests center around deep learning architectures, machine learning, NLP, image processing, and federated reinforcement learning for secure data transmission in wireless sensor networks. She has been involved in several international collaborations and industrial projects, including data-driven solutions for HepsiBurada and AndMe in Turkey, where she developed large-scale AI-based recommendation and forecasting systems. Dr. Seyedeh Tina Sefati’s technical skill set includes advanced proficiency in Python, TensorFlow, PyTorch, CNN, LSTM, GANs, and Transformers, demonstrating her ability to bridge theoretical concepts with real-world applications. Her research excellence is reflected in publications in Scopus and IEEE-indexed journals such as The Journal of Supercomputing and Mathematics. She is a recognized member of professional organizations such as IEEE and ACM and has received honors for her research contributions in deep learning and anomaly detection.
Featured Publications
-
Sefati, S. T., Razavi, S. N., & Salehpour, P. (2025). Enhancing autoencoder models for multivariate time series anomaly detection: The role of noise and data amount. The Journal of Supercomputing, 81(4), 559. (2 citations)
-
Sefati, S. T., Feizi-Derakhshi, M. R., & Razavi, S. N. (2016). Improvement of Persian spam filtering by game theory. International Journal of Advanced Computer Science and Applications, 7(6). (1 citation)
-
Sefati, S. S., Sefati, S. T., Nazir, S., Farkhady, R. Z., & Obreja, S. G. (2025). Federated reinforcement learning with hybrid optimization for secure and reliable data transmission in wireless sensor networks (WSNs). Mathematics, 13(19), 1–37.
-
Sefati, S. T., Razavi, S. N. (2024). Hybrid deep learning approach for intelligent anomaly detection in IoT sensor data. IEEE Internet of Things Journal. (3 citations)
-
Sefati, S. T., Salehpour, P. (2023). GAN-based synthetic data generation for anomaly detection in multivariate time series. Expert Systems with Applications. (4 citations)
-
Sefati, S. T., Feizi-Derakhshi, M. R. (2022). Game-theoretic optimization in distributed deep learning systems. Applied Intelligence. (2 citations)
-
Sefati, S. T., Nazir, S. (2021). Deep learning-based adaptive framework for real-time sensor data analysis. IEEE Access. (3 citations)