Opening speech by Alexander Klimchik - the head of the Center for technologies in robotics and mechatronics components, and Professor Stefano Nolfi - coorganizer, the head of the Laboratory of Autonomous Robotics and Artificial Life, Institute of Cognitive Science and Technology, National Research Council (CNR), Italy

Syllabus:
- A robot in all our homes
- Language learning in babies and robots
- Developmental robotics definition and principles
- Embodied Language learning
- Learning abstract words
- Trust and Theory of Mind
- Spiking and neuromorphic models
- Deep learning for robotics

Syllabus
- Evolving robots
- Embodiment
- Situatedness
- Behavior and cognition as complex dynamical systems
- Swarm Robotics
- Long-term adaptation and open-ended evolution

Machine learning-based computer vision is becoming an integral part of robotics due to its highly accurate results. It provides the vision to understand the surroundings and classify the objects that enable driverless cars on the road and unmanned aerial vehicles (i.e., drones) to bring enormous opportunities. In this lecture, the objective is to gain in-depth knowledge of convolutional neural networks for the classification and semantic segmentation task. It will cover from basics to modern deep learning models which are based on convolutional neural networks. At the end of the lecture, a home learning task will be provided that enabled the participant to use such models for solving problems related to classification and semantic segmentation task.

The lecture will begin with an introduction to Generative and Discriminative Models. Next, we will learn about an Autoencoder, followed by a detailed discussion on two of the state-of-the-art generative models: Variational Autoencoder (VAE) and Generative Adversarial Networks. Finally, we will discuss an application of VAE to the problem of Zero-shot Anomaly Detection.