Pinned Loading

1. GPT-2GPT-2Public

   PyTorch implementation of a GPT-2-inspired transformer language model, trained on English (Shakespeare) and Hebrew corpora, with attention visualizations and evaluation.

   Python 1

2. Denoising-Diffusion-Probabilistic-ModelDenoising-Diffusion-Probabilistic-ModelPublic

   PyTorch implementation of the Denoising Diffusion Probabilistic Model (DDPM) with DDIM sampling and classifier-free guidance, based on Ho et al.'s paper, evaluated on the MNIST dataset.

   Python 5

3. bert-lora-adapter-finetuningbert-lora-adapter-finetuningPublic

   Empirical study of LoRA and adapter-based parameter-efficient fine-tuning of BERT on CoLA.

   Python 1

4. Deep-Reinforcement-Learning-MinigridDeep-Reinforcement-Learning-MinigridPublic

   Deep RL agents for the MiniGrid 6 Rooms. Implements DQN, Double DQN, Huber Loss, Soft Updates, PER, and Actor-Critic. Best model solves 90% of episodes with improved training stability.

   Python 1

5. Auto-Encoding-Variational-BayesAuto-Encoding-Variational-BayesPublic

   PyTorch implementation of the Variational Autoencoder (VAE) model from the Kingma & Welling paper 'Auto-Encoding Variational Bayes', evaluated on MNIST and Fashion MNIST datasets.

   Python

6. RL-FlappyBirdRL-FlappyBirdPublic

   Reinforcement Learning agent that plays Flappy Bird using Q-Learning and SARSA, implemented from scratch. Includes a custom environment, tabular Q-table updates, and performance visualization.

   Jupyter Notebook 1