• Hi!
    I'm Mohammad

    I love AI and my childhood's dream to do anything better with AI is going to happen with my hands.

    View CV

  • I am a
    Computer Engineer

    I am interested in Computer Vision. I am also interested in building agents using Deep Reinforcement Learning for games.

    View Programming Skills
    Summary

About Me

Who Am I?

Hi I'm Mohammad Doosti Lakhani

I am a bachelor degree student of computer engineering at the university of Guilan.
I started my career in AI by studying about Evolutionary Algorithms, like Genetics and I have implemented a paper about using genetics to optimize Multi-Depot Vehicle Routing Problem.
Then, I started learning about machine learning and deep learning on my own through self-study using online videos from different universities. And now I am researching about my ideas.

I love teaching and had many presentations about different topics in almost all undergraduate courses. And Also, I usually contribute to Open Source community couple of times a week.

Furthermore, I love communicating with people to know more and exchange information. This helps me a lot to understand better.

Finally, Reinforcement Learning is the top-1 topic I want to do some research.


Focused on

Visual Computing

Deep
Learning

Reinforcement Learning

Evolutionary Algorithms

Currently focusing on master's thesis about

Multi-Objective Optimization using
First-Order Approximation

Github Status

Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Mon Wed Fri
Contributions in the last year 193 total Mar 15, 2019 – Mar 14, 2020
Longest streak 4 days October 10 – October 13
Current streak 1 day March 14 – March 14
What I do?

My Major Collaborative Skills

Comprehensive Documentations

Capable of providing adequate and transparent docs to alleviate maintenance and further usage.

Presentation

I have had a multitude number of presentations about almost all courses of the curriculum particularly in Artificial Intelligence.

Capable of learning new technologies

Widely enthusiastic about acquiring new skills and capable of adapting to a new circumstances.

Mathematics

I am eager in learning mathematics such as calculus, algebra, matrix theory, etc.

Team Work

Conferring ideas with team members to aggregate all the potential upsides.

My Speciality

My Skills

Please open the boxes to see more specific information.
Education

Education

3rd rank among 2014 computer engineering department students
GPA 93% (18.64 / 20).

Download Untranslated Transcript

Teacher Assistant Experiences

  • Advanced Programming (Head TA) Instructor: Dr. Mirroshandel
    Fall 2018
  • Algorithm Design Instructor: Dr. Shakeri
    Spring 2017
  • Algorithm Design (Head TA) Instructor: Dr. Shakeri
    Fall 2018
  • Computational Intelligence (Head TA) Instructor: Dr. Shakeri
    Spring 2018
Major Courses:
  • Advanced Programming
  • Data Structure
  • Algorithm Design
  • Calculus
  • Statistics
  • Signals & Systems
  • Microelectronics
  • Digital Electronic
  • Computer Vision
  • Artificial Intelligence
  • Computational Intelligence
  • Computer Aided Design (VHDL)
  • Classic NLP
  • Robotics
  • etc .

Awards:

  • Exceptional Talent of Department of Computer Engineering - Aug 2019

  • Full Scholarship of B.Sc Degree - Aug 2015

  • Accepted as M.Sc Degree Exceptional Talent of Department of Computer Engineering With Full Scholarship
    Iran University of Science and Technology - Aug 2019

  • Machine Learning by Stanford Home Page
  • SuperDataScience - Machine Learning Home Page
  • National Research University Higher School of Economics - Natural Language Processing Home Page
  • Neural Networks and Deep Learning by deeplearning.ai on Coursera Home Page
  • The Hong Kong University of Science and Technology - Full Stack Web and Multiplatform Mobile App Development Home Page
  • And some others courses about Android, Web Dev,SQL, etc.

In this school, professors had presentations about ANN, CNN, GAN and etc. Hands on assignments were with python and Keras.

See source page

1st place in the high school with GPA 98% (19.60 / 20)

1st place in the middle school with GPA 100% (20 /20)

1st place in the elementary school with GPA 100% (20 / 20)

Experience

Project Experience

Implementation of "Deep Context-Aware Descreening and Rescreening of Halftone Images" paper via PyTorch. Nov 2018- Present

This project pertains automated Descreening process. Descreening is the task that we try to reconstruct the halftoned image (which is the mandatory process to interact images with printers, scanners, monitors, etc) meanwhile reducing the amount of data loss.
wikipedia/halftone - Paper

The authors, have not published any code about this paper.. So this implementation is the first one and it is fully in PyTorch.
The implementation can be divided into below separate projects:

  • CoarseNet: Modified version of U-Net architecture introduced in "Convolutional Networks for Biomedical Image Segmentation" to work as a low-pass filter by remove halftone patterns.
  • DetailsNet: A deep CNN generator and two discriminators which are trained simultaneously to improve image quality.
  • EdgeNet: A simple CNN model to extract canny edge features as it is necessary as part of the end-to-end learning procedure.
  • ObjectNet: Modified version of "Pyramid Scene Parsing Network" to only return 25 major classified segments instead of 150 to be adapted to halftoning process.
  • Halftoning-Algorithms: Implementation of some of the halftone algorithms provided in most recent digital color haltoning books to prepare data as the input of the whole project.
  • Places365-Preprocessing: A custom and extendable implementation of Dataset abstract class in PyTorch to handle lazy loading of a huge data functionality via utilizing CPU for preprocessing and GPU for training.

Optimized Multi-Depot Vehicle Routing Problem Oct 2017- Dec 2017

MDVRP is a multi-objective optimization task that the goal is to assign a number of vehicles which are distributed in multi depots in search to the customers meanwhile minimizing the number of car used and distance traveled regarding some constraints such as vehicle weight threshold.

At first, I attempted implementing the Using Genetic Algorithms for Multi-depot Vehicle Routing paper, but I noticed that the algorithm could be refined in many aspects, so I applied the desired modification in mutation and replacement so that I achieved better results in term of metrics provided in the original paper.

This project was part of the Computational Intelligence course. Github Repository

Coursera Machine Learning with Python Dec 2018-Nov 2019

In this project, I implemented all assignments of coursera machine learning course by Andrew Ng in python and using native libraries (no octave/matlab to python libraries). I have done all mandatory and optional assignments in vectorized form for sake of optimization. has been done.

Furthermore, I have rewrite the whole documentation with some improvements such as emphasizing on primary sections and generating illustrations dynamically.

Adaptive K-MCI Jan 2018-Present (Sacrificed for greater ideas)

Optimizing Modified Cohort Intelligence via Fuzzy hyper-parameter tuning incorporating clustering using K-Means
The goal is to define a fuzzy system for the hyper-parameters of the k-MCI evolutionary to help it to obtain better results.

Read

Recent Blog

HTML5 Bootstrap Template by colorlib.com
April 14, 2018 | Web Design

Renovating National Gallery

Separated they live in Bookmarksgrove right at the coast of the Semantics, a large language ocean.

HTML5 Bootstrap Template by colorlib.com
April 14, 2018 | Web Design

Wordpress for a Beginner

Separated they live in Bookmarksgrove right at the coast of the Semantics, a large language ocean.

HTML5 Bootstrap Template by colorlib.com
April 14, 2018 | Inspiration

Make website from scratch

Separated they live in Bookmarksgrove right at the coast of the Semantics, a large language ocean.

Get in Touch

Contact

This from uses third-party tools to send your message (Security Issue).
So if you have any concern about it, send your message directly to my EMAIL.