My name is Mohammad KHalooei ('محمد خالوئی' in Persian). I am an artificial intelligence researcher and also fascinated by doing scientific research in Deep learning & Big data fields of academia. And being an enthusiastic entrepreneur at the same time is interestingly intriguing.
By now, I'm a Ph.D. candidate of Amirkabir University of Technology (Tehran Polytechnic) under supervision of Prof. Mohammad Mehdi Homayounpour and Dr. Maryam Amirmazlaghani. My research field of academia is Adversarial Machine learning and Robustness of Deep Neural Networks.
In this paper, we propose a novel self-supervised representation learning by taking advantage of a neighborhood-relational encoding (NRE) among the training data. Conventional unsupervised learning methods only focused on training deep networks to understand the primitive characteristics of the visual data, mainly to be able to reconstruct the data from a latent space. Different from the previous work, NRE aims at preserving the local neighborhood structure on the data manifold. Therefore, it is less sensitive to outliers.
( arxiv) - ( theCVF)
This research inspired by the success of generative adversarial networks for training deep models in unsupervised and semi-supervised settings, we (4 colleague) propose an end-to-end architecture for one-class classification. Our architecture is composed of two deep networks, each of which trained by competing with each other while collaborating to understand the underlying concept in the target class, and then classify the testing samples. One network works as the novelty detector, while the other supports it by enhancing the inlier samples and distorting the outliers. The intuition is that the separability of the enhanced inliers and distorted outliers is much better than deciding on the original samples. The proposed framework applies to different related applications of anomaly and outlier detection in images and videos.
( arxiv) - ( theCVF) - ( IEEE) - ( GitHub)
Optimization is the selection of a best element (with regard to some criterion) from some set of available alternatives. Convex optimization is a subfield of mathematical optimization that studies the problem of minimizing convex functions over convex sets. Machine learning algorithms use optimization all the time. We minimize loss, or error, or maximize some kind of score functions. Gradient descent is the "hello world" optimization algorithm covered on probably any machine learning course. This course was taught by Dr. Amirmazlaghani.
AAIC is stands for Amirkabir Artificial Intelligence Competitions which was held in Tehran, IRAN. This is the forth experience which was held in Tehran and organized with one of the most technology based university of IRAN - Amirkabir University of Technology (Tehran Polytechnic). The fourth mission of AAIC competition had four leagues such as : Face recognition, Object recognition from few training examples, Stock market prediction, Persian named entity recognition and Voice command recognition field. The CEO of the forth mission of AAIC was Prof. Safabakhsh and Dr. Nickabadi.
Pattern Recognition Letters is a peer-reviewed scientific journal that is published by North Holland, an imprint of Elsevier, on behalf of the International Association for Pattern Recognition. The journal produces 16 issues per year covering research on pattern recognition. A certificate of outstanding contribution reviewer in November 2018 was awarded to me.
Statistical Machine Learning (SML in a brief) is a second graduate level course in advanced machine learning, assuming students have taken Machine Learning and Intermediate Statistics. This course was taught by Dr. Nickabadi.
Most parts of my researches is devoted to the newest Deep learning topics, and multimedia user behavior analysis.
Core ML is the machine learning framework used across Apple products, including Siri, Camera, and QuickType. CoreML is integrated machine learning models into your app. Core ML is the foundation for domain-specific frameworks and functionality. Core ML supports Vision for image analysis, Natural Language for natural language processing, and GameplayKit for evaluating learned decision trees. Core ML itself builds on top of low-level primitives like Accelerate and BNNS, as well as Metal Performance Shaders. The machine learning stack Core ML is optimized for on-device performance, which minimizes memory footprint and power consumption. Running strictly on the device ensures the privacy of user data and guarantees that your app remains functional and responsive when a network connection is unavailable.
Machine and Vision Intelligence was part of the IAPR programs and it was organized by “Computer Vision, Pattern recognition and machine Learning Italian Association” (CVPL), affiliated to International Association for Pattern Recognition (IAPR) with support of Technical Committees TC-03 (Neural Networks and Computational Intelligence) and TC-12 (Multimedia and Visual Information Systems). VSIMAC was organized by Prof. Alfredo Petrosino, Prof. Francesco Camastra and Dr. Fabio Narducci.
Most of projects have some analysis part in different data type (i.e. location, sensor, numerical, image, video and sound).
Most parts of my research focus is on Deep learning online courses (MOOCs) . Also, we provide the first deep learning lectures in IRAN and most of our focusable notes are from online and hot topic resources.
Most part of my research focus is on Big data online courses (MOOCs) .
Sentiment analysing of WWW social commenting infrastructure.
- - Qualified for participating RoboCup2017 competition at Nagoya, japan
- - Participating in IRAN OPEN 2017 at Tehran and won 5'th prize
- - Qualified for participating in RoboCup2016 competition at Leipzig, German
- - Participating in IRAN OPEN 2016 at Tehran and won third prize
- - Participating in IRAN OPEN 2015 at Tehran
- - Participating in AUT Cup 2015 at Amirkabir university
Some of my public presentation files: [I try to gather all presentation files and placed them here! (in near future :))]
I will update this table in near future ...
Some research and industrial projects :