Top tools for Photographers
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Dolore, porro rem quod illo quam, eum alias id, repellendus magni, quas.
I recently graduated with a Bachelor of Computer Engineering from the University of Guilan. During my studies, I never settled for less and constantly looked up ways to improve myself as a student of the game. To take full advantage of my potential, I decided to tackle various aspects such as research, sports, and student work. Happily, I successfully assessed and supervised about 300 fellow students during my TAships, represented my country on an international stage, and accomplished multiple research projects. As an Artificial Intelligence researcher who began his journey in AI in his fourth semester, I reached out to researchers around the globe to broaden my studies and figure out my ultimate passion by tackling problems from different origins. I am proud to conclude that I will be after applying my humble knowledge of AI on the human brain to play a small part in aiding mankind towards a brighter future.
Reza Khan Mohammadi
22
Rasht, Iran
B.Sc in Computer Engineering
GPA: 18.35/20 (3.86/4)
Pre-University Diploma in Mathematics
GPA: 18.19/20 (3.8/4)
Diploma in Mathematics and Physics Discipline
GPA: 18.33/20 (3.8/4)
As an assignment, I had the chance to generally present Natural Language Processing and its applications, besides my papers.
You can find presented slides here.
I presented my team’s final project for FISU president and other chief members alongside 130-members audience. The presentation had to be concluded in just a minute, in which all materials we learnt during the academy and all ideas we implemented in our team project had to be reviewed.
I was a guest speeker at Rsaht School of AI’s fourth meetup session and presented the predefined learning path and shared my experiences with the other participants.
You can find presented slides here.
As the main speaker of the event, I held a conference to celebrate International Day of University Sports (IDUS) in which I took the chance to familiarize students with volunteering, FISU structures and to share my experiences.
You can find presented slides here.
In my first semester, I had a conference about fundamentals of computer engineering and to share my programming experiences with other students.
The prenatal identification of fetal gender is a day-to-day linical task that is mostly carried out using Ultrasounds. In his work, we tackle the problem of fetus gender identification using acoustical sounds of the heart as recorded by PCG ignals, and extend upon previous studies in CVD detection o the task presented in this paper. We established a baseline by experimenting with different denoising methods and eature extraction techniques to derive meaningful information from these signals to aid the modeling of fetus gender using classical ML and DL algorithms. Our proposed ensemble method significantly improved the baseline by correctly classifying 91% of the FPCG signals among unseen subjects. Our results motivate further work in robust denoising techniques, and the development of end-to-end systems for modeling heart activity and FGI.
Having curated a parallel corpus of prose and their equivalent poems, we introduce a novel Neural Machine Translation (NMT) approach to translate prose to ancient Persian poetry using transformer-based Language Models in an extremely low-resource setting. More specifically, we trained a Transformer model from scratch to obtain initial translations and pretrained different variations of BERT to obtain final translations. To address the challenge of using masked language modelling under poeticness criteria, we heuristically joined the two models and generated valid poems in terms of automatic and human assessments. (A sample of one of our pretrained models on Huggingface)
In this research project we gathered two datasets: Persian COVID-19 Articles and PerSICK. The former consists of Healthcare articles related to COVID-19. The latter is the first Persian dataset used to measure Semantic Textual Similarity. The primary contribution of this paper is a novel Document Ranking approach which utilizez a ranking and re-ranking stage using BM25, TF-IDF, and BERT.
In this research project, We studied linguistical and sociolinguistical differences of male/female text based on their word choices and proposed a novel approach of style transfer in gender domain. An other interesting novelty of this research was the Multilingualism of our method, which unlike other approaches, is able to be applied on multiple languages. All in all, the outcome of this research was the proposal of PGST, a Polyglot Gender Style Transfer.
In this Project, we have introduced the first instance of Text Style Transfer in Persian Language. This approach heavily relies on Greedy Search algorithm, Neural Networks, and Word Embeddings.
EEG Signal Processing
In this work, we have been experimenting with different EEG signal datasets such as Epilepsy and BCI Competition IV to start my journey in the field of Brain-Computer Interfaces. I have been implementing different filters, techniques, and Machine Learning models from scratch to fundamentally tackle this field.
Fetal Heartbeat Sound Classification
In this project, we tackled a Heartbeat sound dataset to experiment with different features which make it possible for us to classify fetuses' heartbeats by their gender
Persian Abstractive Summerization
In this work, we gathered a large-scale Persian dataset for
abstractive summarization and explored multiple approaches for such manner. During the process not
only I have faced with the difficulties of applying some of the most popular abstractive summarization
techniques, but also we are working on a brand new method of summarization for obtaining better
results in Persian to end up publishing it as a research paper.
Bilingual Image Captioning
In this research project, We are gathering a Persian Image Captioning dataset and investigating different effects of Multi-task Learning to generate better Persian captions.
Polyglot Text Style Transfer
In this research project, We studied linguistical and sociolinguistical
differences of male/female text based on their word choices and proposed a novel approach which of
style transfer in gender domain. An other interesting novelty of this research was the Multilingualism
of our method, which unlike other approaches, is able to be applied on multiple languages.
• Persian Text Style Transfer
In this Project, we have introduced the first instance of Text Style Transfer in Persian Language. This approach heavily relies on greedy search, neural networks, and word embeddings.
Document ranking using Transformers
In this research project, we gathered two datasets: Persian COVID-19 Articles and PerSICK. The former consists of Healthcare articles related to COVID-19. The latter is the first Persian dataset used to measure Semantic Textual Similarity. The primary contribution of this paper is a novel document ranking approach which utilizez a ranking and re-ranking stage using BM25, TF-IDF, and BERT.
My responsibilities were to design and assess language and speech processing assignments and final projects. Besides, I designed academic tutorials for a class of 40 students in which I developed hands-on Deep NLP projects.
My responsibilities were to design and assess multiple Deep Learning/Machine Learning final projects and single-handedly guide a class of more than 50 students through. Additionally, I designed tutorials in which I reviewed popular Data Science libraries (Numpy and Pandas) and thought how to develop Neural Networks using Pytorch.
My responsibilities were to design and assess 5 series of assignments among three other TAs in a class of more than 70 students
During weekly sessions that I held with students, I taught them multiple complex data structures (graphs, trees, hash tables and...) in both Java and Python. Moreover, alongside two other TAs, I designed and assessed 5 projects during the semester for a class of more than 50 students
Alongside weekly sessions in which we focused on developing object-orientation, GUI design and socket programming in Java, I designed and assessed 7 projects in a team with 2 other TAs for a class of more than 70 students.
My responsibilities were to design and assess 7 series of assignments and assess students’ final project in a team of 3 other TAs for a class of more than 70 students.
My only responsibility was to assess student assignments in a class of more than 50 students.
I am responsible for holding AI events and providing free material for those who want to pursue The AI field.
My responsibility is to manage around 300 volunteers in different phases for the whole event.
Served a year as a FISU student Ambassador in my country.
Held multiple national Sport/Cultural event.
Held multiple conferences (main speaker) and specialized workshops in University of Guilan to develop volunteering morale among students.
Held a streetball championship in University of Guilan.
Single-handedly designed both Back-end and Front-end of a web-based system using PHP, HTML, CSS and MySQL to automate the process of storing PC specifications in the faculty of Sport Science and Physical Education
Gathered questionnaires and performed different statistical tests.
May 2019
Sep 2019
Check out my latest blog posts
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Dolore, porro rem quod illo quam, eum alias id, repellendus magni, quas.
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Dolore, porro rem quod illo quam, eum alias id, repellendus magni, quas.
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Dolore, porro rem quod illo quam, eum alias id, repellendus magni, quas.
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Dolore, porro rem quod illo quam, eum alias id, repellendus magni, quas.
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Dolore, porro rem quod illo quam, eum alias id, repellendus magni, quas.
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Dolore, porro rem quod illo quam, eum alias id, repellendus magni, quas.
Feel free to contact me anytimes