Some projects related to Advanced Machine Learning:

• Image Classification. The task is to train a pretrained Convolutional Neural Network (CNN) architecture on a dataset. Dataset: Cats vs Dogs. Architecture: ResNet18
• Object Detection. The task is to test three (3) pretrained models on the test set of a given dataset, comparing the results between the models. Dataset: Coco detection. Models: Faster R-CNN, SSD (Single Shot MultiBox Detector), Retina Net.
• Image Captioning. The task is to train an Image Captioning model using Encoder-Decoder with Attention on a subset of the COCO dataset. Tune the hyperparameter values of the model and for the optimization such that a BLEU score of at least 0.2 is achieved on the test set.
• Spiking Deep Q-Learning