Course

Deep Learning - June 2025

In this course, participants will get acquainted with the basics - theoretical and practical - of neural networks, what problems we encounter, how we solve them and how we ensure the correct operation of our models. Along with this, it will be explored how, with the help of abstraction and combining building blocks, students will be able to build, train and test complex architectures. They will understand the basic principles behind modern research and applications of neural networks, how to follow their principles and how to create their own, original solutions. Participants will look at the cases in which deep neural networks find the greatest application - working with images and natural languages, and will understand how generative models work and how to teach a model to make autonomous decisions.

Topics

Intro
Intro

No description yet
Resources
- Intro
Introduction to deep learning. Basic models
Introduction to deep learning. Basic models

1. Computational graphs
2. Simple models with tensorflow and pytorch
- Low-level API
3. Building neural networks
4. Training and evaluation
5. Regularization

06/22 Sunday - 10:30

Outside - 404
Online - Zoom 15:00
Resources
- Lecture
- Video - Record
Training and improving neural networks
Training and improving neural networks

1. Regularization
2. Bias and variance
- Error analysis
3.Optimization algorithms
4. Hyperparameter tuning
5. Normalization

06/25 Wednesday - 18:00 Outside - 404
06/26 Thursday - 18:00 Online - Zoom
Resources
- Lecture
- Video - Record
Neural networks for images
Neural networks for images

1. Convolutional neural networks
- Operations
- Architectures
2. Generalizations
- ResNet
- 1x1 convolutions, "network-in-network"
3. Object localization

06/28 Saturday - 10:00 Outside 404
06/29 Sunday - 10:00 Online - Zoom
Resources
- Lecture
- Video - Record
Neural networks for language processing
Neural networks for language processing

1. Time-dependent (sequential) models
- Architecture
- Types
2. Improvements
3. Word (token) representations
4. Refinement algorithms
- Attention
-Transformers

07/02 Wednesday - 18:00 Outside - 404
07/03 Thursday - 18:00 Online - Zoom
Resources
- Lecture
Advanced neural network architectures
Advanced neural network architectures

1. Architectures: review
2. Transfer learning
3. Semi-supervised methods
4. Image captioning
- Multimodal deep learning

07/05 Saturday - 10:00 Outside 404
07/06 Sunday - 10:00 Online - Zoom
Resources
- Lecture
Generative models
Generative models

1. Generative models: basics
2. Unsupervised data generation
- Style transfer
- Generating sequences
- Variational autoencoders
3. Advanced generative models
- Reminder: transformers
- Generative adversarial networks
- Stable diffusion
- NeRF

07/09 Wednesday - 18:00 Outside - 404
Resources
- Lecture
Reinforcement learning
Reinforcement learning

1. Problem description
2. Approaches
3. Deep-Q networks
4. AlphaGo
5. "Specification gaming"

07/23 Wednesday - 18:00 Outside - 404
07/24 Thursday - 18:00 Online - Zoom
Resources
- Lecture
Deep learning in Production
Deep learning in Production

Work shop

07/26 Saturday - 10:00 Outside 404
07/27 Sunday - 10:00 Online - Zoom
Resources
- Doc
Exam preparation: end-to-end project
Exam preparation: end-to-end project

Work shop

07/30 Wednesday - 18:00 Online - Zoom
Resources
- Doc

Who is the target audience for the course?

The course is suitable for people with basic programming knowledge (Python language is recommended). Knowledge of mathematics at the level of the 12th grade is mandatory (experience with elements of higher mathematics is also desirable). On the plus side are the completed Data Science and Machine Learning trainings or similar knowledge, such as skills for data collection and analysis, creating and applying machine learning models, using the scientific method for decision-making, building complete projects and solving problems using statistical methods and machine learning algorithms. The required level of English required for the course must be B2.

How to enroll in the course?

To sign up for the course, contact MLC Business School and follow the steps provided. After a successful payment, you will be enrolled in the training.

What is the deadline for enrolling in the course and when does it start?

Enrollment in the course is open until 05 February. The training starts on 15th May 2025. You can find a detailed schedule of classes in the Topics section.

How and where are the classes be held?

You can study online in real time. Immediately after each lesson, you also get access to the lesson recording and learning resources.

What is the date of the exam and what does it include?

The exam will be held online and includes practical exercises. The dates of the exam are 19th and 26th of June 2025.

Do I get a certificate after the exam?

After passing the exam, you acquire a certificate issued by SoftUni if your score is above 70%.

How much is the course fee and what does it include?

Online

400 USD

Online training in real-time

Lifetime access to lesson recordings and learning content

Help from a mentor in understanding the learning materials

Access to a closed Facebook group with all other course participants

Taking a regular exam and receiving a certificate

SoftUni gives you a 100% guarantee of the quality of this course. The most important thing for us is that you acquire the necessary skills and knowledge. In the event that the training fails to fulfil your expectations, we guarantee a full refund of the amount you have paid. You can receive a refund until the fifth lesson of the course (30 May).