MATLAB and Simulink Seminars

Demystifying Deep Learning: A Practical Approach in MATLAB

Welle7 (at Bern main railway station), Bern, Switzerland


Session 2: Hands-on Workshop is fully booked - for waiting list, please send an email to

Are you new to Deep Learning and want to learn how to apply these techniques in your work?

Deep Learning achieves human-like accuracy for many tasks considered algorithmically unsolvable with traditional Machine Learning. It is frequently used to develop applications such as face recognition, automated driving, and image classification.

The main tasks are to assemble large data sets, create a neural network, to train, visualize, and evaluate different models, using specialized hardware - often requiring unique programming knowledge. These tasks are frequently even more challenging because of the complex theory behind them.

We will demonstrate new MATLAB features that simplify these tasks and eliminate the low-level programming. In doing so, we will demystify practical knowledge of the domain of Deep Learning.

This free full-day event is divided into two sessions: A seminar-style morning and a hands-on workshop in the afternoon. Please feel free to register for the session that suits you best.

Please note: Seats for the afternoon session are limited, and will be confirmed in a separate email. For participation at the workshop, please bring along your laptop. Wi-Fi, software and access to computing resources (cloud, GPU) will be provided (no installation required).


  • Introduction to Deep Learning and its use cases
  • Perform classification and pixel-level semantic segmentation on images
  • Import training data sets from networks such as GoogLeNet and ResNet
  • Import and use pre-trained models from TensorFlow and Caffe for Transfer Learning
  • Automate manual effort required to label ground truth
  • Build a deep network that can classify your own handwritten digits
  • Speed up network training with parallel computing on multiple CPUs/GPUs
  • Learn how to use CUDA code generation technology to accelerate inference performance on a GPU

Who Should Attend

Engineers, scientists and anybody interested in the development of Deep Learning applications.


Time Title




How to build an autonomous anything with AI


Demystifying Deep Learning (Part 1)

  • Why Deep Learning
  • MNIST: The “Hello, World!” of computer vision
  • Transfer learning with CNNs


Coffee Break


Demystifying Deep Learning (Part 2)

  • Semantic segmentation
  • Ground truth labeling of datasets
  • Everything else in Deep Learning…


Demo Presentation


Lunch Break

Hands-On Workshop


  • Introduction
  • Deep Learning for image classification


Coffee Break


  • Deep Learning for time series
  • Automatic CUDA code generation from MATLAB code


Conclusion and Q&A

16 :30


Product Focus

Registration closed