Research

Machine Learning Applications in Wireless Communication Networks

Goal: The project aims to detect user anomalies and inference patterns in wireless networks, targeting to improve the data rates of communication systems.

Thesis: Machine Learning Applications in Wireless Communication Networks: Interference Detection and Authentication Aspects. 

All the slides:

Device_Authentication

Machine Learning Applications in Magnetic Domain Pattern Images

In order to extract parameters from and/or generate new images of magnetization patterns using micromagnetic simulation data, the images of simulated magnetization patterns will be input into Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) for training.


Benefit: An opportunity to create domains that have out-of-range parameters (you can’t execute in micromagnetic simulation). 

Generative_Adversarial_Network

Decoding Motor imagery-based EEG Signals using Machine Learning

Brain-computer interfaces represent a life-changing possibility for disabled patients to regain control of their lives, for example through highly functional prosthetics or movement devices. However, everyday movements, like walking, and the increased brain activity it causes pose significant difficulty for BCI decoders. 

This paper proposes a method to clean EEG data and create a decoder that is resistant to walking-induced artifacts so left and right-hand motor imagery can be classified. We first perform Artifact Subspace Reconstruction to remove major artifacts, then bandpass in the alpha and mu bands to find physiologically relevant frequencies, Common Average Reference Filter to improve spatial resolution, extract trials and samples, find the mean percent difference in power spectrum density between resting and motor imagery to best identify event-related desynchronization, perform fisher score feature selection to reduce dimensionality, and finally train and cross-validate our model offline and test online with evidence accumulation.


The paper is attached under publications. 

Some of the code for this work is available at https://github.com/zkhodzhaev/ML_brain_computer_interface

EEG data before and after ASR processing. The upper subplot shows raw and the lower illustrates the filtered signal where the spike is not present.

The topological plot of the grand averaged mu power from the last 0.5s of the task trials, with and without spatial filtering.

Feature Extraction/Selection

Fischer scores measuring statistical significance for the first experimental run 1 to 4 in session one for subject 41. The y-axis shows Fischer score values ranging from 8 to 30 Hz with ∆2. The x-axis shows the number of channels from one to fifty-five. The color represents the discriminant power of the Fischer score with higher scores indicating features that better differentiate the conditions/classes:

Feature extraction for testing the ML model

Rethinking Fair and Stable Representation Learning

Machine Learning, particularly in the context of Graph Neural Networks (GNNs), is being deployed across various applications, including safety-critical domains. 


While the accuracy of these systems remains a critical consideration, it is equally important to examine them through the lenses of fairness and stability. Prior research has introduced a unified framework to learn stable and fair node representations. 


In this work, we revisit and refine this framework, presenting a simplified version that demonstrates superior performance in terms of both accuracy and fairness/stability metrics. 


Furthermore, we introduce a set of straightforward baseline methods that prove to be effective and competitively perform, often surpassing the performance of previous state-of-the-art (SOTA) approaches. 


The paper is attached under publications. 


The code for our work is available at https://tinyurl.com/AMLFinalCode 

and specifically to my contribution (MLP): https://github.com/zkhodzhaev/representation_learning

Spike-Timing-Dependent Plasticity (STDP) Neuromorphic Devices 

Coming soon...

Unconventional Computing 

Skyrmion stability and movement under different temperature, current density, and physical constraints. 

Skyrmion: Material Growth

Coming..

Novel Materials: Dynamics of Hopfions

Study of resonant spin dynamics of topological spin textures.

Using micromagnetic simulations, e.g., MuMax3 and OOMMF to find resonant spin dynamics of a three-dimensional topological spin texture hopfion in different chiral magnets and identifying the ground state spin configuration of hopfions, effects of anisotropies, geometric confinements, and demagnetizing fields. Calculation of the resonance frequencies and spin-wave modes of spin precession dynamics under multiple magnetic fields.

Novel Materials: Antiferromagnetic Material Switching

NiO Antiferromagnetic Switching:

You cracked it!

Electric Field Effects on the Chocolate Flow

The electric field was applied along the direction of the flow of chocolate and the effects of the electric field on temperature, its gradient, and pressure were investigated. 

PID was used for controlling temperature and setting them. Solid-state-relays were used to heat the system and nitrogen gas was used for pressuring the system. An electric field was created using Cu plates. 

The graph of chocolate weight versus time was graphed in real-time.

Electric_field_effect_on_chocolate_flow