Tejas Bana

About Me

I am a graduate student pursuing a Master of Science in Electrical and Computer Engineering at Carnegie Mellon University. My background spans both research and industry, with a strong emphasis on generative models, privacy-preserving machine learning, and on-device machine learning. I thrive on the challenges of applied machine learning problems, relishing the opportunity to apply my expertise to real-world scenarios.

In my recent industry experience, I had the privilege of developing deep learning systems for understanding chemical reactions and molecular representations. Among my most memorable projects is the time I spent collaborating with the Indian Army to create dynamic camouflage pattern generation software.

Courses & Projects

• Learning-Based Image Synthesis 16-726
  Main Project: GANs to Understand How the Human Brain Makes Sense of Natural Scenes
  GAN based: GAN photo editing Neural Style Transfer When Cats Meet GANs
  Other: Gradient Domain Fusion Colorizing the Prokudin-Gorskii Photo Collection
  

• Understanding and Critiquing Generative Computer Vision 16-895
  Exploring nuances of modern day generative models, understand how they work so well, and develop reasonable critiques of their capabilities to understand how they can be made even better and applicable for diverse downstream research themes.

• On-Device Machine Learning 11-767
  Learning to build efficient machine learning systems through quantization, pruning, and neural architecture search for optimized power consumption and memory utilization.

• Neural Signal Processing 18-698
  Learning neural decoding, firing rate estimation, neural system characterization, sensorimotor control, spike sorting, and field potential analysis.

• Advanced NLP 11-711
  

• Large Language Models Methods and Applications 11-667
  

• Introduction to Machine Learning 18-661
  

• Computational Models of Neural Systems 15-883
  

Publications

• Robust Recovery of Adversarial Samples
  Tejas Bana, Jatan Loya, Siddhant Kulkarni
  ICML-2021 Workshop on AML
  Paper

• Privacy-Preserving Keystroke Analysis using Fully Homomorphic Encryption & Differential Privacy
  Jatan Loya, Tejas Bana
  International Conference on Cyberworlds (CW 2021)
  Paper

• Vit - Inception - GAN for Image Colourizing
  Tejas Bana, Jatan Loya, Siddhant Kulkarni
  The Conference on Machine Learning, Optimization and Data science (LOD)
  Paper