azaddeihim Goto Github PK

acopf-ppo-gnn

cartpole-dqn

Uses a deep Q-learning based agent to solve CartPole

exoplanet-detection

Exoplanet detection using highly imbalanced keplar mission dataset.

face-generation

Used generative adversarial networks to generate realistic images of human faces. Implemented metrics to quantitatively evaluate the quality of generated images including generative adversarial metric, and Fréchet Inception distance.

frequency-control-maddpg

Uses multi-agent deep deterministic policy gradient to balance demand and generation using frequency control in a distributed generation system.

hmdb51-action-recognition

Created several models to classify actions in HMDB51 dataset. Used models consisting of convolutional and pooling layers for feature extraction followed by recurrent neural networks for classification. Devised novel two-stream architecture capturing spatial and temporal features of input videos, in conjunction with Mask R-CNN for background removal, with bidirectional long-short term memory for classification, earning 80% accuracy on the test set.

maze-solving-q-learning

Uses a Q-learning based agent to solve a maze.

sttre