Get in Touch

Course Outline

Deep Learning vs Machine Learning vs Other Methods

  • Appropriate scenarios for Deep Learning
  • Constraints and limitations of Deep Learning
  • Evaluating accuracy and cost across different methods

Methods Overview

  • Networks and Layers
  • Forward / Backward: The core computations in layered compositional models
  • Loss: How the learning objective is defined by the loss
  • Solver: The mechanism coordinating model optimization
  • Layer Catalogue: The layer as the basic unit of modeling and computation
  • Convolution

Methods and Models

  • Backpropagation and modular models
  • Logsum module
  • RBF Net
  • MAP/MLE loss
  • Parameter Space Transforms
  • Convolutional Module
  • Gradient-Based Learning
  • Energy functions for inference
  • Learning objectives
  • PCA; NLL
  • Latent Variable Models
  • Probabilistic LVM
  • Loss Function
  • Object detection using Fast R-CNN
  • Sequence modeling with LSTMs and Vision + Language integration via LRCN
  • Pixelwise prediction with FCNs
  • Framework design and future directions

Tools

  • Caffe
  • Tensorflow
  • R
  • Matlab
  • Others

Requirements

A foundational understanding of any programming language is required. While prior familiarity with Machine Learning is not mandatory, it is advantageous.

 21 Hours

Number of participants


Price per participant

Testimonials (3)

Upcoming Courses

Related Categories