Loading Events

« All Events

Introduction to Deep Learning

November 7 - November 8

Deep Learning Essentials is a 3-days course designed to provide a comprehensive understanding of the fundamentals of deep learning. The course covers common deep learning architectures included in machine learning tasks such as supervised and unsupervised learning. Starting from a brief description of gradient-based learning algorithms, several common deep architectures are going to be covered, including:

  • Multi-layer Perceptrons and Deep Architectures
  • Deep Autoencoders
  • Convolutional Neural Networks
  • Recurrent Neural Networks
  • Generative Adversarial Networks

The participants will get familiar with implementing the state-of-the-art deep learning models on cutting-edge open source deep learning technology, primarily Tensorflow and Keras. It will cover practical considerations when developing a deep learning application, focusing on tuning parameters, avoiding overfitting, accelerating optimization algorithms and transfer learning. The attendees will have access to cloud-based notebook environments including all necessary technology as well as a list of demonstrative notebooks.


Hands-On Labs:


  • With hands-on labs and demonstrations, the participants will utilize the ecosystem tools in order to:
  • Prepare a deep-learning environment for advanced machine learning tasks
  • Understand deep-learning architectures
  • Write simple programs to perform common machine learning tasks using open source deep learning libraries
  • Write simple programs to perform computer vision tasks including object recognition/segmentation and emotion detection.
  • Write simple programs to perform natural language processing tasks such as sentiment analysis and named entity recognition
  • Observe how complex NLP applications are built in practice
  • Observe how algorithms can learn with trial and error
  • Observe how algorithms can actually synthesize data


Course Prerequisites:


The attendees are expected to be able to write simple programs in Python, but the amount of programming is minimal.


Course Coverage:


Machine Learning Essentials:

  1. Learning from Data
  2. Common Machine Learning Tasks
  3. Machine Learning Pipelines:
    1. Data Preparation
    2. Learning
    3. Evaluation
  4. Linear Models & Logistic Regression
  5. Learning with Stochastic Gradient Descent
  6. Overfitting & Regularization
  7. Transfer Learning


Introduction to Deep Learning

  1. Multi-layer Perceptron and Backpropagation
  2. Accelerated Optimization Methods
  3. Autodifferentiation with Tensorflow
  4. Deep Architectures
  5. Classification and Regression with Deep Models


Learning Representations

  1. Autoencoders
  2. word2vec
  3. Lab: Learning Semantic Relationships from Text Corpora


Learning from Structured Modalities

  1. Convolutional Neural Networks
  2. Lab: Learning Abstract Local Structures from Images
  3. Lab: Object Recognition
  4. Demonstration: Emotion Detection


Learning from Sequences

  1. Recurrent Neural Networks
  2. Long Short-Term Memory
  3. Lab: Language Modeling
  4. Demonstration: Sequence-to-sequence classification


Generative Models

  1. Generative Adversarial Networks
  2. Lab: Algorithms that synthesize data
  3. Demonstration: Generating artwork


Istanbul Venue
Istanbul, 34345 Turkey
+90 212 217 63 88