Do you want to know the secret behind the AlphaGo agent that beat the 18-time world champion in the Go game in 2017?

Zaka is offering a 2-day ONLINE Reinforcement Learning (RL) hands-on crash course happening on Friday, 7th August from 15h00 to 20h00 (GMT+3) and on Saturday, 8th August from 10h00 to 18h00 (GMT+3). 

This crash course introduces the basics of RL and teaches the needed ingredients to cast the decision-making processes that we face daily in different fields (engineering, gaming, finance, …) into well-formulated RL problems. 

Throughout the hands-on crash course, you will be able to interact with RL environments and agents in OpenAI Gym and implement algorithms (such as Q-learning) to develop smart RL agents. 

The course also provides an introduction to Deep RL through the implementation of deep Q-networks in RL settings. 

Join us from the comfort of your home, discover the basics of the RL field, and experience building your own algorithms on some of the most commonly used platforms! 

Who can attend? 

This course is introductory and designed for university students, professionals, tech enthusiasts, career transitioner, or anyone who is interested in a practical dive into the fundamentals of RL.

All you need to have is

• Basic programming background (preferably in Python)
• Basic knowledge of probability theory
  
  

Meet your instructor Julia El Zini

She is pursuing her PhD in computer engineering from AUB and has 4 years of experience in several machine learning fields. She worked on projects that include reinforcement learning, advanced parallelization, optimization techniques, generative adversarial networks, attention networks, and graph convolution neural networks.

Learning objectives

1- Get introduced to Reinforcement Learning (RL), its definition and requirements

2- Learn the basis of RL formulations

• Get introduced to the Markov Decision Process (MDP)
• Identify tricks on how to provide good formulation 

3- Define the policy and link it to the goal of RL

• Policies in deterministic environments
• Policies in stochastic environments

4- Get introduced to different RL methods

5- Familiarize with OpenAI Gym

• Create an RL environment
• Interact with an agent

6- Implement the basic Q-learning algorithm

7- Learn and implement Deep Q-learning

• Introduce deep neural networks
• Implement Deep Q-networks in RL settings

What Else?

Certificate

All attendees will receive a certificate from ZAKA upon completing the crash course. Completion is conditioned upon attending all the sessions

How Does it Work?

You will receive a Zoom meeting link a day ahead of the scheduled event. You can easily access the online session through the link. We will be communicating more detailed instructions via email, as well as any downloadable resources and notebook solutions.