In this ongoing research project at UMBC, which is under the supervision of Prof. Sanjay Purushotham , the focus is on generating high-quality PCG signals. I utilized the Physionet Moody Challenge 2022 dataset. The shared image represents a preliminary result of the quality of the synthetic PCG data generated through our methodology. The project's codebase will be made available after the work is published.
During my research assistance ship at the KUSIS AI lab at Koc University, I worked under the supervision of Prof. Baris Akgun on various DL and RL based projects. The latest work focuses on proposing a high-level decision-making mechanism, rooted in offline RL methods, for discretionary lane-change behavior on highways.The above gif shows the performance of our offline methodology on a Python coded Environment.
This is my Masters thesis project which was under the supervision of Prof. Baris Akgun. This work proposes a data-driven methodology to create decision aids for refinery operators. These aids use machine learning models trained on historical data to generate soft-alarms that alert operators of potential undesirable events within a specific timeframe.The study also explores an online adaptive thresholding method to fine-tune the precision-recall trade-off, enhancing the decision-making process.
About
Hi there! Welcome to my homepage!
I am Ainaz Jamshidi, A Phd stdent at UMBC, My research areas generally lie in Data Science, Machine Learning, Deep Learning, Reinforcement Learning , Computer Vision, Time series Forecasting. My current research projects are predominantly centered on synthetic data generation through GANs and Diffusion models for health care applications.
Also, I am well experienced in Python, R programming, PyTorch, Tensor flow, etc.
GPA: 4.0 out of 4.0 (Full merit-based scholarship from Al-Ghurair Foundation for Education).
Relevant Coursework: Advances in Deep Learning, Deep Unsupervised Learning, Deep Learning and Computer Vision, Introduction to Machine Learning, Natural Language Processing, Algorithms and Complexity, Data Structure & Algorithms, and Software Engineering Analysis and Design.
Work Experience: Undergraduate TA in advanced programming in JAVA, and peer-to-peer programming tutor.
Koc University
Istanbul, Turkey
Masters in Computer Science
GPA: 3.83 out of 4.0 (Full merit-based scholarship from KUSIS AI Lab ).
Relevant Coursework: Advances in Deep Learning, Deep Unsupervised Learning, Deep Learning and Computer Vision, Introduction to Machine Learning, Reinforcement learning, Algorithms and Complexity, Data Structure & Algorithms
Work Experience: TA in advanced programming in Python, Data Structure, Introduction to Artificial Intelligence, and peer-to-peer programming tutor.
Amirkabir University of Technology
Tehran, Iran
Bachelor in Electrical Engineering
GPA: 3.80 out of 4.0
Relevant Coursework: Signals & System, Bio signal processing, Adaptive signal processing, DSP.
Work Experience: Data Science internship at Atiyeh clinical neuroscience center.
Achievements
Offered Senior Data Scientist Role at Trendyol-The largest online shopping platform in Turkey.
2023
Awarded the KUSIS AI scholarship at Koc University.
2022
Kaggle Expert, received 1 silver and 8 bronze medals.
2021
Exceptional talent award from Amirkabir University.
2015
Top 0.1% of participants in the nationwide university entrance exam in Iran.
2013
Professional experience
KUIS AI LAB Research Assistance
Istanbul, Turkey
Supervised by Prof.Baris Akgun
September 2018 - 2023
Worked on several projects based on Deep Learning (DL) and Reinforcement Learning (RL), including:
Operator Decision Aid Design via Multi-Dimensional Time-Series Event Prediction: A Hydrocracking Unit Application,
Learning Autonomous Discretionary Lane Change behaviour utilizing Offline Reinforcement Learning
CareX
California, USA
Remote Part-Time Machine Learning Engineer
Nov 2021 - March 2022
Designed a pipeline to extract PPG signals from the videos recorded from the finger tip by smart phones’ camera.
Designed and implemented Machine learning pipelines for blood pressure estima- tion using Pytorch and Keras.
Private Tutoring
London, England
Remote
September 2021 - 2023
Delivered +400 hours of private tutoring in introductory Python programming, Advanced Python programming, Algorithm and complexity.
Data Scientist
Tehran, Iran
Supervised by Dr.Golnaz Baghdadi
June 2017 - June 2019
Designed and implemented a pipeline for denoising EEG signals based on ICA algorithm using Matlab and WinEEG.
Designed and implemented a new two back task with positive and negative feed- back in c# programming language. This implementation was employed, by the clinic, in their studies for a long time.
Research on cognitive neuroscience, designing and programming cognitive tasks.
INDEPENDENT PROJECTS & SKILLS
Generation of High Quality Synthetic Phonocardiogram Signals Using Generative Adversarial Networks and Diffusion Models.
Learning Autonomous Discretionary Lane Change behaviour Based on Offline Reinforcement Learning.
year: 2022
Submitted to IEEE Transactions on Intelligent Transportation Systems.
Technical Skills & Tools:
Programming Languages And Tools: Python, C/C++, Java, Julia, MYSQL.
Advanced algorithms, Data Structure, Machine learning, Deep learning, Reinforcement Learning, and the basics of Web development, and Android development.
Unix, LaTex, Web scraping, Git, etc.
Contact
Do not hesitate to reach out if you have full-time positions, internship opportunities, or just want a simple chat.
Feel free to contact me via the form below or at ainazj1@umbc.edu