Biography

I am a graduate student in Electrical and Computer Engineering at University of California Los Angeles

My broad areas of interests are: Fair and Responsible AI, Statistical Machine Learning, Digital Signal Processing, Medical Imaging, Audio Processing, Probability and Matrix Theory. I have worked on various projects on these topics and the list can be found on my resume.

Apart from my academic work, I enjoy playing and writing music. I am an experienced guitarist and my greatest satisfaction is a song well made

• LORAKI: Reconstruction of Undersampled k-space Data using Scan-Specific Autocalibrated Recurrent Neural Networks
  Tae Hyung Kim, Pratyush Garg, Justin Haldar

  International Society for Magnetic Resonance in Medicine (ISMRM), May 2019, Montreal, QC, Canada.

  PDF

• Joint Mobile Sink Scheduling and Data Aggregation in Wireless Sensor Networks using Q-Learning
  Surender Redhu, Pratyush Garg, Rajesh M. Hegde

  IEEE International Conference on Acoustics, Speech and Signal Processing, 2018, Alberta, Canada.

  PDF

• LORAKI: Reconstruction of Undersampled k-space Data using Scan-Specific Autocalibrated Recurrent Neural Networks

  This approach reconstructs the image from undersampled k-space using a scan-specifically trained network. We use the phase and space constraints inherent to k-space along with calibration data as the training set. Results indicate better performance than peers and further improvement possible with optimizations.

• Joint Mobile Sink Scheduling and Data Aggregation in Wireless Sensor Networks using Q-Learning

  The collection of data from sensor nodes using mobile sinks becomes difficult when the network is asynchronous since the energy of the sink is also limited. To solve this problem, we implemented an algorithm that uses deep Q networks to optimize the collection of data with the mobile sink energy.

• Scaling Mixed Membership Stochastic Blockmodels (MMSB) to Large Datasets
• Multiobjective Optimization using Pareto Descent and Multiple GD Algorithm
• Mutiresolution Analysis of EEG Signals using basis derived from Orthogonal Chebyshev Polynomials

• pratyushg@g.ucla.edu
• pratyushgarg11@gmail.com
• +1 (469)-837-9674
• 512 Veteran Avenue, Los Angeles.