Lots of buzzwords, lots of unicorns, but we will make it practical and lead you in your first steps to the magical place of data science, specifically, deep learning. We will talk a bit about the theory behind it, but mostly, we will demonstrate the variety of common techniques and usages. You will acquire the fundamental knowledge and tools to implement deep learning architecture by yourself.
The workshop is for anyone who has at least 2 years of experience with programming (preferably some knowledge of python) , logic orientation, and an outstanding will to learn.
In the workshop, we will use python, sklearn, tensorflow and keras.
This is a complex topic, you should work hard and be very focused during those weeks ^_^
Each workshop day is built from lecturing and hands-on practice.
Don’t forget to bring your laptop!
Practice makes perfect – at the each workshop day you will have lots of home exercise to help you examine all you’ve learned .
What differs AI from other kind of algorithms? What’s the relation between AI and Machine Learning? How did it all start and what is Deep Learning? In this lecture we will cover the buzz words, history and general notions around the topic of AI from a technical perspective.
We’ll gain a deep intuitive understanding of the machine learning field. Start with covering the main concepts and tools of machine learning, both theoretically and practically.
This is the fundamentals for the understanding of how to use Deep Neural Networks.
Among the main topics: what is the prediction pipeline, feature engineering, linear & logistic regression, learning tasks and measurement metrics.
History and motivation
Computer vision / image processing
Generative adversarial networks (GANs)
Natural language processing
Multimodal methods combining image and text Video, audio, speech Game playing
Training and optimizers
MNIST and CIFAR10 datasets
Convolutional neural networks
Preprocessing and data augmentation
Common architectures for image recognition
Overview of approaches for other computer vision tasks
Anomaly detection (optional)
Using deep neural networks for image classification.
The basic building blocks.
Tips and tricks in image classification .
Object detection basics, classification with localisation.
Different types of Object detection networks – SSD , YOLO.
Learn how to show this object classifying pixels of the image segmentation.
Different concepts of semantic segmentation vs instance segmentation.
Technological advances in artificial intelligence allow to take photos of real life objects and automatically create 3D models out of them. This is going to change the way a 3D designer works, allowing for much more efficiency and time saving.
We will see a neural network that takes as input a 2D image and automatically a 3D model, using an encoding-decoding architecture. A ResNet based encoder is trained to encode the image into a z-vector with inherent 3D features and a decoder which is actually a boolean classifier is trained to create a 3D model from the z-vector. The reconstruction can happen in any voxel resolution, without retraining the network. Also we will discuss some of the challenges with 3D modelling and ML, we will present cool implementations of ML in the visualization, texture analysis, 3D modeling and other relevant subjects.
How to Learn from Little Data
One-shot classification for a small labeled image dataset.
We’ll also go over the architecture of its inspiration.
Deep convolutional GAN
Generative Adversarial Networks.
We’re going to use a Deep Convolutional GAN to generate images of pokemons. Architecture and implementation.
Summarize & what’s next.
Meet the bigger circle, those you donated and helped to!