Deep Learning Bootcamp 2017

Europe/Berlin
Seminar Room 2 (Center for Systems Biology Dresden)

Seminar Room 2

Center for Systems Biology Dresden

Pfotenhauerstr. 108 01307 Dresden Germany
Florian Jug, Peter Steinbach
Description

This hands-on course will take you from 0 to 100 in Deep Learning with Keras. Our aim is to teach the fundamentals of deep learning with Convolutional Neural Networks (CNN) based on modern techniques using the Keras API and the Tensorflow backend. By the end participants will know how to build deep learning models, how to train them, what to avoid during training, what to check during training and how to perform model inference, especially for image based problems. We hope participants will then go out and apply these methods to their own problems and use cases.

The core curriculum is planned from Tuesday (August 22) to Thursday night (August 24) to take place at the MPI CBG campus, Dresden, Germany. As the agenda is currently being prepared please check-in from time to time. There is the latent plan to have voluntary session before (introduction to python) and after the workshop (hackday to work on the challenges you have). This is why the workshop currently blocks the entire week.

All participants are expected to bring their laptop. During the workshop, a uniform access to GPU-enabled workstations or servers will be provided that hold the software stack used. Thus, your laptop is not required to hold a mobile GPU or alike. All participants are expected to have a solid understanding of fundamentals of linear algebra as well as programming.

The workshop admission fee amounts to € 100 per participant. Every successfull applicant is required to bring a poster to the workshop that describes their current scientific challenge that they would like to solve with Deep Learning. Posters have to be sent in 1 week prior to the workshop. 

As part of the workshop, Google engineers will join us to demonstrate using the Google cloud for deep learning with tensorflow. This event is open to anyone interested, but admission is only allowed after registraton here: https://indico.mpi-cbg.de/event/68/.

We have a slack channel for asynchronous and offline communication. Feel free to join in through this link. We are also on twitter, if you want to chime in use  #DLBC17 .

Slides
    • 13:00 18:00
      Introduction to python Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany
      Convener: Peter Steinbach
      etherpad
      etherpad at the end of the day
      Jupyter Notebook Service
      self-study
    • 09:00 12:00
      Deep Machine Learning Fundamentals Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany

      Machine learning basics:
      - Bias-Variance Tradeoff
      + Cross-Validation
      + Training, Validation and Test sets

      • The image classification problem:

        • Linear approach
        • k-neareast neighbour
      • Data driven approach:

        • Loss function
        • Regularization
        • Optimization
      • Back propagation

    • 12:00 13:00
      Lunch 1h Canteen (Max Planck Institute of Molecular Cell Biology and Genetics)

      Canteen

      Max Planck Institute of Molecular Cell Biology and Genetics

      Pfotenhauerstr. 108 01307 Dresden

      The costs for lunch will be covered for all registered participants that bring their name batch to the canteen.

    • 13:00 14:00
      Image Classification with NVIDIA DIGITS 1h Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany
      If you want to sign up for the qwiklab exercises yourself, please create an account here: https://nvlabs.qwiklab.com/ This lab shows you how to leverage deep neural networks (DNN) - specifically convolutional neural networks (CNN) - within the deep learning workflow to solve a real-world image classification problem using NVIDIA DIGITS on top of the Caffe framework and the MNIST hand-written digits dataset. In this lab, you will learn how to: - Architect a Deep Neural Network to run on a GPU - Manage the process of data preparation, model definition, model training and troubleshooting - Use validation data to test and try different strategies for improving model performance On completion of this lab, you will be able to use NVIDIA DIGITS to architect, train, evaluate and enhance the accuracy of CNNs on your own image classification application.
      cloud platform
    • 14:00 14:30
      Q&A Image Classification with NVIDIA DIGITS 30m Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany
    • 14:30 18:00
      Deep Learning Fundamentals continued Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany
      jupyter notebooks
      kaggle ottogroup dataset
      zalando dataset
    • 09:00 12:00
      Deep Learning Intermediate Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany

      Network architecture
      - Activation functions
      - Data preprocessing
      - Weight initialization:
      + Batch normalization
      - Convolutions
      - Pooling

      CIFAR10 notebook
      MNIST jupyter notebook
    • 12:00 13:00
      Lunch 1h Canteen ()

      Canteen

      The costs for lunch will be covered for all registered participants that bring their name batch to the canteen.

    • 13:00 14:30
      Cray Tools + FAQ 1h 30m Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany
      Speaker: Mr. Findling Andreas (Cray)
    • 14:30 18:00
      Deep Learning Intermediate continued Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany
    • 09:00 12:30
      Deep Learning Advanced Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany
      • Best practices:
        • Adam
        • Learning rates
      • Transfer learning
      • Object detection
      • Object segmentation
      • Generative Models (DGAN)
      • Text-based use cases (RNNs)
      GAN python script
      Transferlearning notebook
    • 12:30 13:30
      Lunch 1h Canteen (Max Planck Institute of Molecular Cell Biology and Genetics)

      Canteen

      Max Planck Institute of Molecular Cell Biology and Genetics

      The costs for lunch will be covered for all registered participants that bring their name batch to the canteen.

    • 13:30 18:30
      Deep Learning with Tensorflow on Google Compute Engine Big Auditorium (Max Planck Institute of Molecular Cell Biology and Genetics)

      Big Auditorium

      Max Planck Institute of Molecular Cell Biology and Genetics

      This session will
      include how to run Tensorflow on Google Cloud Platform, including on GPUs
      as well as demonstrating some other Deep Learning benefits of using the
      Google Cloud Platform. At the end, there will be a Q&A session.

      For more details, see:
      https://indico.mpi-cbg.de/event/68/overview

      Convener: Peter Steinbach
      notebooks as a zip file
      slides
    • 19:30 22:00
      Workshop Dinner Seminar Room 2

      Seminar Room 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany

      Dinner at a close-by Biergarten on a self-pay basis

    • 09:00 16:00
      Hackday/Ask-the-expert SR 2

      SR 2

      Center for Systems Biology Dresden

      Pfotenhauerstr. 108 01307 Dresden Germany

      Full-day event where participants are asked to prepare a poster which illustrates their current challenge that they would like to solve with DL

      final etherpad of the week
    • 12:00 13:00
      Lunch 1h Canteen ()

      Canteen

      The costs for lunch will be covered for all registered participants that bring their name batch to the canteen.

Your browser is out of date!

Update your browser to view this website correctly. Update my browser now

×