Dr. Muhammad Saif-ur-Rehman

Development of advanced self-supervised machine learning algorithms. In particular, my projects involved the following methodologies. 1. Development of adaptive, self-supervised and self-organising machine learning algorithm for neuroscience applications. 2. Investigating the ability of domain adaption of deep learning algorithms. 3. Development of feature extraction and data augmentation algorithms for EEG signals.

Applications of machine learning algorithms & statistical modeling in brain-computer interface application. In particular, my projects involved the following methodologies. 1. characterization of cortical signatures of motor imagery signal using deep learning algorithms. 2. Development of statistical in conjunction with machine learning approaches for the analysis of non-stationary , high dimensional, and noisy neural data. 3. Development of decoding framework of neural signal with integrated GUI.

Anomaly detection in the electricity consumption of more than 200 stores. I have developed a statistical method based on Gaussian mixture model (GMM) to detect occurrence of anomaly.

I have investigated the vulnerability of deep learning algorithms (Conventional Neural Networks) against certain type of inputs called adversarial inputs. Now a days, this has become a very important and one of the most well researched field in machine learning, known as Generative adversarial networks (GANs).

I was responsible for analysis of linguistics data, features extraction and apply machine learning for classification

An investigation of intriguing properties of neural networks. In these days, they are called "Generative Adversarial inputs (GANs)"

I was responsible for conducting undergraduate labs for different electronic engineering courses

Implication of machine learning algorithms in brain-computer interface (BCI)

Machine Learning

Communication Enginneering