Skip to content

Latest commit

 

History

History
58 lines (42 loc) · 2.62 KB

NLP and Dialogue Systems-Focused CV.md

File metadata and controls

58 lines (42 loc) · 2.62 KB
Raw

Yasaman Saffari

PhD Candidate in Artificial Intelligence
Kashan University, Iran
Email: yasamansaffarii@gmail.com | Phone: +98 937 755 3756
Location: Tehran, Iran
LinkedIn | Google Scholar | ResearchGate | GitHub

Summary

PhD candidate with a strong focus on Natural Language Processing (NLP) and Task-based Dialogue Systems. Expertise in developing dialogue systems using reinforcement learning, graph embeddings, and attention mechanisms. Proven track record of research publications and practical implementations of state-of-the-art NLP models. Dedicated to advancing the field of NLP through innovative research and collaboration.

Education

PhD in Artificial Intelligence
Kashan University, Iran
2020 - Present

  • Advisor: Dr. Javad Salimi Sartakhti
  • Research Focus: Dialogue Systems, Reinforcement Learning, State Representation Learning, NLP, Graph Representation Learning, Attention Mechanisms
  • GPA: 3.81/4.0

Master of Arts in Intelligent Simulator Design
Tabriz Islamic Art University, Iran
2017 - 2019

  • Thesis: "Investigating the Effective Variables of Speech Phonology Using Deep Neural Networks for Voice Cloning in Dubbing"
  • GPA: 3.79/4.0

Research Experience

PhD Researcher
Dialogue Systems and Reinforcement Learning
Kashan University, Iran
2020 - Present

  • Developed and optimized dialogue systems using reinforcement learning algorithms (DQN variants).
  • Focused on graph-based state representation learning and bias mitigation in NLP applications.
  • Conducted research on hierarchical attention mechanisms for enhancing dialogue policy learning.

Publications

  • Y. Saffari, J. Salimi Sartakhti (2023). "Actor Double Critic Architecture for Dialogue Systems," Journal of Electrical and Computer Engineering Innovations (JECEI), 11(2).
  • Y. Saffari (Under Review). "CODACAN: Distributional Actor Critic with Hierarchical Attention-based State Representation for Dialogue Policy Learning," Knowledge-Based Systems.
  • Additional papers under review focusing on graph-structural state embeddings for dialogue systems.

Skills

  • Programming and Tools: Python, TensorFlow (Keras), PyTorch
  • NLP Techniques: Reinforcement Learning, Deep Learning, Graph Embedding, Attention Mechanisms, Transformers
  • Tools: Overleaf, VSCode, Git

Awards and Honors

  • Full Fund Scholarship with Quota for Brilliant Talents, Kashan University, Iran (2020 - Present)
  • First Rank in Global Game Jam, Sharif University of Technology (2020)

Languages

  • Persian (Native), English (Fluent)