I’mMahan Veisi, a Computer Engineering undergrad (graduatingJan 2026) atShahid Beheshti University.
I work at the intersection ofGenerative AI (diffusion models),Medical Imaging, andComputational Biology.

📫Find me: 🌐Website · 🎓Google Scholar · 💼LinkedIn · ✉️Email ·▶️YouTube

PinnedLoading

1. LSTMppo-DRL-StockTraderLSTMppo-DRL-StockTraderPublic

   Deep Reinforcement Learning (DRL) stock trading system with LSTM-PPO, integrating technical indicators and FinBERT-based news sentiment analysis.

   Jupyter Notebook 29 4

2. Latent-Diffusion-MNIST-DDPM-using-AutoencoderLatent-Diffusion-MNIST-DDPM-using-AutoencoderPublic

   Unlock the potential of latent diffusion models with MNIST! 🚀 Dive into reconstructing and generating digits using cutting-edge techniques like Autoencoders with Channel Attention Blocks and DDPMs.…

   Jupyter Notebook 17

3. From-Chaos-to-Clarity-Denoising-Images-with-UNet-and-GANsFrom-Chaos-to-Clarity-Denoising-Images-with-UNet-and-GANsPublic

   ✨ Dive into image denoising magic! This project uses Attention U-Net and PatchGAN to tackle noise types like low Gaussian and salt-and-pepper noise. Perfect for computer vision, deep learning, and …

   Jupyter Notebook 11

4. VAE-MNIST-Variable-Latent-Size-Reconstruction-and-VisualizationVAE-MNIST-Variable-Latent-Size-Reconstruction-and-VisualizationPublic

   Dive into the world of Variational Autoencoders (VAEs) with MNIST! 🎨✨ Explore variable latent sizes (2, 4, 16) to see how they affect reconstruction, latent space visualizations, and performance me…

   Jupyter Notebook 10

5. FashionTaggerFashionTaggerPublic

   Upload your fashion images and get detailed labels, such as category, color, and season, predicted by my AI model using an ensemble of CNN architectures.

   Jupyter Notebook 14

6. RL_practicesRL_practicesPublic

   Collection of my Reinforcement Learning (RL) practices including DQN, D3QN, and Adaptive Gamma, applied to the Lunar Lander and CartPole environments. 🚀🕹️

   Jupyter Notebook 17