UT Introduction to Neural Networks and Deep Learning Course

Projects of Introduction to Neural Networks and Deep Learning Course - Fall 2022 - University of Tehran

CA1 - (Simple Neural Networks)

• Q1: McCulloch-Pitts neural model - 2-Bit Binary Multiplier
• Q2: AdalLine and MadaLine - Binary Classification
• Q3: Restricted Boltzmann Machine - Collaborative Filtering
• Q4: Multi Layer Perceptron - House Price Prediction

CA2 - (CNN)

• Q1: Effects of Varying Resolution on Performance of CNN based Image Classification
• Q2: CNN Model for Image Classification on MNIST and Fashion-MNIST Dataset

CA3 - (Region based CNNs)

• Q1: Transfer Learning - AlexNet
• Q2: Delving into High-Quality Synthetic Face Occlusion Segmentation Datasets
• Q3: YOLOv6: Real-Time Object Detection

CA4 - (CNN-RNN)

• Q1: Air‑pollution prediction in smart city, deep learning approach
• Q2: Fake News Detection: A hybrid CNN-RNN based deep learning approach

CA5 - (Transformers)

• Q1: Implementation of BERT
• Q2: BEIT: BERT Pre-Training of Image Transformers

CA6 - (GAN)

• Q1: Implementation Deep Convolutional GAN
• Q2: implementation Auxiliary Classifier GAN

Extra

• Q1: Credit Card Fraud Detection Using Autoencoders
• Q3: A recognition model for handwritten PersianArabic