This course provides a broad introduction to the basic concept of the classical neural networks (NN) and its current evolution to deep learning (DL) technology. The primary goal of this course is to introduce the well-known deep learning architectures and their applications in defense and security for object detection, identification, verification, action recognition, scene understanding and biometrics using a single modality or multimodality sensor information. This course will describe the history of neural networks and its progress to current deep learning technology. It covers several DL architectures such the classical multi-layer feed forward neural networks, convolutional neural networks (CNN), restricted Boltzmann machines (RBM), auto-encoders and recurrent neural networks such as long term short memory (LSTM). Use of deep learning architectures for feature extraction and classification will be described and demonstrated. Examples of popular CNN-based architectures such as AlexNet, VGGNet, GooGleNet (inception modules), ResNet, DeepFace, Highway Networks, FractalNet and their applications to defense and security will be discussed. Advanced architectures such as Siamese deep networks, coupled neural networks, auto-encoders, fusion of multiple CNNs and their applications to object verification and classification will also be covered.
- Identify the fundamental concepts of neural networks and deep learning.
- Understand the major differences between neural network and current deep learning architectures.
- Explain the stochastic gradient descent algorithm to train deep learning networks with different regularizations methods.
- Describe the popular CNN-based architectures (i.e., AlexNet, VGGNet, GooGleNet, ResNet).
- Compare the relative merits of various deep learning architectures, MLP, CNN, RBM and LSTM.
- Formulate CNN and auto-encoders for feature extraction.
- Demonstrate the use of deep learning framework for object, face, pedestrian detection, pose estimation and face identification.
- Differentiate between Siamese and coupled deep learning architectures and their use for object verification and identification.
- Design multiple deep learning architectures for multi-view face identification and multimodal biometrics applications.