COMPUTER ENGINEERING STUDENT

• CAMP chair Technical University of Munich
• Email: arshiahemmat6@gmail.com
• LinkedIn: ArshiaHemmat

Education

Technical University of Munich (TUM)

• VISITOR STUDENT RESEARCHER IN CAMP
  • Advisor: Prof. Nassir Navab
  • Location: Munich Germany
  • Duration: Jun. 2023 - PRESENT

Sharif University of Technology (SUT)

• STUDENT RESEARCHER IN RIML
  • Advisor: Prof. Mohammad Hossein Rohban
  • Location: Tehran IRAN
  • Duration: Nov. 2022 - PRESENT

University of Isfahan (UI)

• B.S. IN COMPUTER ENGINEERING
  • GPA: 3.83 out of 4
  • GPA of last four semesters: 4 out of 4
  • Thesis topic: Graph Fusion in Multi-modal Networks with Zero-shot Evaluation
  • Location: Isfahan IRAN
  • Duration: Sept. 2019 - PRESENT

Research Interests

• Zero-shot Learning
• Vision-linguistic Models
• Explainable AI
• VQA
• Self-supervised Learning
• Geometric Deep Learning

Publication

• CLIP Exhibits Improved Compositional Generalization Through Representation Disentanglement.
  • Authors: Reza Abbasi, Amirarshia Hemmat, Mahdi Samiei, Mohammad Hossein Rohban, Mahdieh Soleymani Baghshah
  • Conference: ICLR 2024

Research Experiences

• 3D Vessel Segmentation Using Geometric Approaches
  Advisor: Prof. Nassir Navab, Technical University of Munich
  Focusing on leveraging geometric deep learning techniques for the segmentation of blood vessels in 3D imaging, this project aims to improve diagnostic accuracy and treatment planning in medical practice.

• Counterfactual Modelling Using Vision-Linguistic Models
  Advisor: Prof. Nassir Navab, Technical University of Munich
  Investigating the generation of counterfactual scenarios through the integration of visual and linguistic data, this research holds promise for enhancing AI interpretability and reliability across various applications, including autonomous driving and medical diagnostics.

• Vision-Linguistic Models
  Advisor: Prof. Mohammad Hossein Rohban, Sharif University of Technology
  This research explores the integration of visual perception and language processing in AI systems, aiming to advance machine understanding and interpretation of complex visual-textual information.

• Zero-shot Learning in Medical Domain
  Advisor: Dr. Peyman Adibi, University of Isfahan
  Developing models for identifying medical conditions unseen during training, this work is poised to revolutionize diagnostic methodologies by mitigating the reliance on extensive labeled datasets.

• Text Summarization Using Graph Neural Network
  Advisor: Dr. Hamidreza Baradaran, University of Isfahan
  Employing graph neural networks for efficient text summarization, this project seeks to enhance the coherence and relevance of automated summaries, facilitating improved information retrieval and understanding in voluminous text data.

Honors & Awards

• TUM Summer Research Fellowship, Technical University of Munich, 2023
• Top 5% Academic Ranking, University of Isfahan, 2023
• 2nd place in National Data Science Competition, Isfahan Iran, 2022
• 22nd in ICPC Asia Tehran, Tehran Iran, 2020
• 3rd place in Poytek business summer school, Isfahan Iran, 2020
• Math Kangaroo diploma, Diploma of Honor in International Kangaroo Mathematics Competition, Tehran Iran, 2015

Teaching Assistant Roles

• Courses: Deep Learning, Machine Learning, Knowledge Systems, Medical Image Processing, Advanced Programming, Linear Algebra, Artificial Intelligence, Engineering Probability and Statistics, Fundamental Programming, Computer Architecture, Object Oriented Analysis and Design, Fundamentals of Robotics, Data structures

Associations

• BigData Lab, Board Member, UI, 2022 - 2023
• Rasta, Chairman & Head of Development Core Board Member, SUT, 2023 - PRESENT
• Community for Artificial Intelligence, Chairman, UI, 2020 - PRESENT
• TEDx University of Isfahan, Organizer, Co-organizer, Advisor, UI, 2020 - 2023

Skills

• Machine Learning Tools: Proficient in Pytorch, NetworkX, Scikit-learn, OpenCV, Pandas, Numpy, MathPlotLib. Familiar with Tensorflow, Keras, Bokeh.
• Programming Languages: Proficient in Python, C++. Familiar with Java, Matlab, VHDL.
• Languages: Native Persian, Advanced English.