My 30-Hour Deep Learning Course: Demo in New York

An intrepid Deep Learner leveraging gradient descent to identify the global minimum

On Saturdays from March 3rd through to April 7th, I’ll be offering an in-classroom-only Deep Learning course at the NYC Data Science Academy. All details, including the full curriculum, are available here. If you live in the area and are keen to experience a high-level demonstration of the course content — as well as to ask me any questions you might have — that’s coming up on the evening of February 20th.

Guess this city

The course is an introduction to artificial neural networks that brings high-level theory to life with interactive labs featuring TensorFlow and Keras, the two Python libraries currently gaining the most popularity. The content of the course is the basis for my textbook, Deep Learning Illustrated, which is being published by Pearson and will appear on bookshelves later this year.

Over five weekends, essential theory will be covered in a way that provides an intuitive understanding of Deep Learning’s underlying foundations. Paired with hands-on code demos in Jupyter notebooks as well as strategies for overcoming common pitfalls, this foundational knowledge will empower individuals with no previous understanding of neural networks to build production-ready Deep Learning applications across all the major contemporary classes:

  • Convolutional Networks used for, e.g., Machine Vision
  • Recurrent Neural Networks, especially LSTMs, used for, e.g., Natural Language Processing and Time Series Analysis
  • Generative Adversarial Networks for creating novel visual art
  • Deep Reinforcement Learning

In addition, I’ll guide you through the creation of your own Deep Learning project from conception through to completion. Together, we’ll engage in the following five project stages:

  1. broad ideation on problems you’re keen to solve and the availability of relevant data sets
  2. formulation of project specifics, including candidate modelling strategies
  3. assessing the performance of your Deep Learning model relative to benchmark approaches
  4. improving your model performance via neural-network architecture modification and hyperparameter tuning
  5. effectively presenting your project and its novel results

As detailed on my testimonials page, students on the previous iteration of my course, held in late 2017, found it tremendously valuable. Mahipal, a software development director at KPMG, indicated the course “was exactly what I hoped it would be. It gave me a strong foundation in all of the core deep learning concepts… it motivated me to pivot my career into deep learning.” Meanwhile Richard, an electrical engineer and former investment banker, said that he “had a ton of fun with the class… Jon was able to illustrate the complex concepts with super-easy-to-understand visualizations.”

I’m now excited to get started with the next cohort! I can’t wait to share my unbridled enthusiasm for Deep Learning and to see the projects that everyone develops.

Jon Krohn supporting Deep Reinforcement Learning’s foundational concepts

Jon Krohn is Chief Data Scientist at the machine-learning startup untapt. He presents an acclaimed series of tutorials on artificial neural networks, including Deep Learning with TensorFlow LiveLessons. He also teaches his deep learning curriculum in-classroom at the NYC Data Science Academy. Jon holds a doctorate in neuroscience from Oxford University and has been publishing on machine learning in leading academic journals since 2010. His forthcoming book, Deep Learning Illustrated, is being published on Pearson’s Addison-Wesley imprint and will be distributed in 2018.

Source: Deep Learning on Medium