Deep Learning & Reinforcement Learning Training
This five-day hands-on Deep learning and Reinforcement learning course is designed for all those seeking a better understanding and knowledge of the major technology trends driving data science.
Attendees will get a clear understanding of the core machine learning concepts, as well as Deep learning and Reinforcement learning techniques and engineering solutions for daily usage. You will go through the complete process of building machine learning systems, from data understanding to modeling.
During hands-on labs, accompanying each theoretical unit, you will gain experience in building and applying deep neural networks and machine learning models with such widely used frameworks as Keras, TensorFlow, and scikit-learn.
At the end of the course, the participants will be able to design working scripts that can be used as a basis for creating algorithms to address business-specific challenges.
Deep Learning & Reinforcement Learning Training
$3400-3900
Buy ticketWhy enroll
- Get particular insights of Reinforcement learning (RL) concept, and techniques with a focus on its practical use.
- Learn about Deep learning (DL) concept and techniques sufficient to be used in reinforcement learning.
- By the end you'll have a deep understanding of the technology with practical experience of its usage in real life cases.
{{eventName}}
Select location
When
Number of participants
Total price
${{commonPriceString}}
The price rises closer to the type of training. Have time to buy now!From - To
The course could be tailored to suit your needs and objectives. It can also be delivered on your premises if preferred.
Price from
${{privatePriceString}}
From - To
The course could be tailored to suit your needs and objectives. It can also be delivered on your premises if preferred.
Price from
${{privatePriceString}}
{{trainingDescription.description}}
{{trainingDuration.description}}
This course assumes that you have some basic knowledge of Machine Learning concepts. It will be useful to get familiarised with calculus, linear algebra and applied statistics to understand the theory behind the algorithms.
All practice is done in Python 3. We highly recommend to use Anaconda distribution and have it installed prior to the training sessions. (Google Colaboratory is an alternative option). It's good to have a laptop to take part in the lab-style part of the course.
Deep Learning part will require Tensorflow (latest stable version is 1.14) and Keras packages to be installed. Tensorflow has GPU version that uses video cards for significant performance boost, mostly suitable to use with NVIDIA cards and requires proper drivers for this version to work.
Reinforcement Learning part will also require OpenAI Gym.
Assumptions: some additional tools might be required. The exact list will be determined and shared with attendees in advance.
Who should attend
The course was designed for:
- Software Engineers and Data Scientists who already have basic practical experience in Machine Learning and looking to implement more advanced algorithms for problem solving.
Training program
-
This is an introduction to the Deep Learning methods for Machine Learning tasks. During this day, we’ll look at surprisingly strong machine learning techniques that have become really popular recently and will cover the following topics:
-
Structure of neural networks, feedforward neural networks
-
A mechanism for learning neural networks
-
Means of neural network learning process control
-
-
Image processing benefitted drastically from Deep Learning. The main architecture for these tasks is Convolutional Neural Network. Topics for the day will include:
-
Image features and representation learning
-
A convolution layer and a deep convolutional network
-
Supporting layers for convolutional neural networks
-
State-of-the-art architectures for image processing
-
Transfer learning and fine tuning
-
-
This day is dedicated to the architecture of neural networks that allow to work with sequential data, most notably, texts. During this day we will cover:
-
Examples of sequential data and related machine learning tasks
-
The vanilla recurrent neural network architecture and its limitations
-
The advanced recurrent neural network layers architecture (LSTM, GRU)
-
-
This day is dedicated to establishing a theoretical base of Reinforcement Learning methods. We’ll also look at the most common solutions for Reinforcement Learning that could be used to address NIWC tasks:
-
Theoretical overview of reinforcement learning task
-
Multi-Armed Bandits (acquiring new knowledge and optimizing decisions based on existing knowledge, balance these tasks to maximize their total value)
-
Markov Decision Processes (mathematical framework for modeling decision making process in situations where outcomes are partly random and partly under the control of a decision maker)
-
Temporal-Difference Methods (reinforcement learning principle that enables online learning from actions directly). Q-learning (method that estimates value of taking an action in different situations)
-
-
Deep Reinforcement Learning combines principles from Reinforcement Learning with Deep Learning. Resulting combination is allowing us to build algorithms that solve complex tasks in different environments.
-
Limitations of basic Reinforcement Learning algorithms (and possible tricks to extend the capabilities for classic methods of Reinforcement Learning)
-
Deep Q-learning (adaptation of Q-learning algorithm to tackle more complex environments)
-
Actor-Critic models (addition to Agent scheme that allows to build more effective algorithms)
-
{{eventName}}
Select location
When
Number of participants
Total price
${{commonPriceString}}
The price rises closer to the type of training. Have time to buy now!From - To
The course could be tailored to suit your needs and objectives. It can also be delivered on your premises if preferred.
Price from
${{privatePriceString}}
From - To
The course could be tailored to suit your needs and objectives. It can also be delivered on your premises if preferred.
Price from
${{privatePriceString}}
Training from authorized Partner
Meet us
Meet the trainers
Kanstantsin Buzanouski is Data Scientist who strives to solve business problems with a defined plan at all stages of the development process.
Kanstantsin’s professional expertise includes:
- considerable hands-on experience in using machine learning and statistical methods in various domains
- solid expertise in solving business problems starting from problem definition up to model fine-tuning and solution deployment
- works with popular frameworks and packages in both R (tidyverse and caret) and Python (pandas/Numpy and scikit-learn)
- profoundly knowledgeable about current trends and approaches to machine learning
Our Attendees
Here is what our customers say about us
