Discount -30% off
who want to start their journey in machine learning from scratch
COURSE ON MACHINE LEARNING AND GENERATIVE AI
For whom?
For whom?
wishing to gain in-demand skills in AI from scratch
beginners, analysts and developers
employees of corporations willing to work on a project in AI
entrepreneurs who want to create a project in AI
The result of the course is a project in a portfolio and practical skills for working with machine learning
/
Real-data portfolio projects and lab work to dive deep into the specifics of working with machine learning
//
Live lectures from leading experts in machine learning
///
Networking in a community of learners and professionals where you can discuss complex topics, share experiences and find answers to questions
Examples of projects
Examples of projects
-
Practice-oriented approachEach unit of the course includes real cases and projects. -
Current program for Q4 2024The course covers the latest trends and innovations in machine learning, including generative models, deep learning, natural language processing, and computer vision. -
Personalized supportThroughout the course, you will have access to mentor support and feedback on your projects. -
Certificate of CompletionUpon successful completion of the course, you will receive a certificate recognizing your knowledge and skills in machine learning.
Teachers
Teachers
-
Oleg BulyginMachine Learning / Deep LearningCodeMind Academy teacher
7 years of experience in managerial positions in research and production companies in the space industry
-
Georgy MilyutinEDA / Mathematics and Statistics
Master of Applied Mathematics and Informatics
- CodeMind Academy instructor.
-
Alexey YaroshenkoDL NLP / GENERATIVE AI: PROMPT ENGINEERING
-
Oleg BulyginMachine Learning / Deep LearningCodeMind Academy teacher
7 years of experience in managerial positions in research and production companies in the space industry
-
Georgy MilyutinEDA / Mathematics and Statistics
Master of Applied Mathematics and Informatics (NIU ITMO)
- Ex-Head of Mathematics at the St. Petersburg branch of the London Gates Education Group
- In 2016-2022, she was a teacher of mathematics and statistics in English for GCSE, IB, SAT, IAL programs
- Author of courses in probability theory and statistics, lecturer at several universities
- CodeMind Academy instructor.
-
Alexey YaroshenkoDL NLP / GENERATIVE AI: PROMPT ENGINEERING
- For the last 3 years in VK, I helped make language models and antifrauds
- Lecturer in ML at HSE IMSH
- Project Mentor at MADE.
program
program
INCLUDED IN THE MODULE:
Mathematical analysis
Probability theory and mathematical statistics
Mathematical analysis
- Gradient descent
Probability theory and mathematical statistics
- Random numbers
- conditional probability
INCLUDED IN THE MODULE:
Basic machine learning
Classification, decision trees, and kNN
Linear regression and classification. MLE, Bias variance decomposition, Regularization, Validation and learning curves
Ensembles of algorithms and the random forest. Bagging. Boosting. Determining the importance of parameters
Feature Engineering
Working with big data
Basic machine learning
- Machine Learning Tasks
- Predictive analytics and quality metrics
- The concepts of model training, application, and validation.
Classification, decision trees, and kNN
- Classification problem formulation
- Metrics
- Decision tree
- The nearest neighbor method
- Model parameter selection and cross-validation
- Application examples and complex cases
- Pros and cons of kNN and trees.
Linear regression and classification. MLE, Bias variance decomposition, Regularization, Validation and learning curves
- Metrics. Why do you need
- Baseline linear regression (LRC)
- Maximum likelihood method
- Linear classifier
- Logistic regression as a linear classifier.
Ensembles of algorithms and the random forest. Bagging. Boosting. Determining the importance of parameters
- Ensembles
- Random forest
- Decision trees vs bagging
- Convert a dataset to a multidimensional representation
- Pros and cons of a random forest
- GBM in machine learning
- Regression loss functions
- Classification functions of losses
- Weights.
Feature Engineering
- Feature selection
- Text, images, geodata, date and time, time series, web, etc.
- Fiche Transformation
- Normalization and change of distribution, interactions, filling in empty values
- Label Encoding
- One-Hot Encoding
- Clustering. K-means.
Working with big data
- SGD
- An online approach to learning.
INCLUDED IN THE MODULE:
Introduction to neural networks
CNN
Introduction to neural networks
- The neuron and the activation function. The simplest neural network
- Neural network training (epochs, mini-batch gradient descent, loss function, cross-entropy loss)
- Batchnorm, dropout, weight regularization and early stopping
- Backpropagation
- Transfer learning
- Learning rate, adaptive LR
- Softmax
- MLP: perceptrons, device.
CNN
- Architecture, principle of operation of convolution
- Layer types
- Filter hyperparameters
- Hyperparameter tuning
- Activation features.
INCLUDED IN THE MODULE:
Introduction to CV
Object detection
Autoencoders
GAN
Introduction to CV
- A brief history of CV development (ImageNet, not neural methods -> CNN -> Transformer).
Object detection
- Введение. Template Matching. Corner Detection
- Edge Detection. Grid Detection. Contour Detection
- Feature Matching. Watershed Algorithm. Custom Seeds with Watershed Algorithm
- MeanShift and CamShift Tracking Theory. OpenCV.
Autoencoders
- Overview
- K-means
- Expectation Maximization
- Linear Autoencoder = Iterative PCA
- The Gaussian Mixture generative
- model
- Binary Autoencoder
- MNIST Autoencoder
- PCA.
GAN
- Motivation
- Generative vs Discriminative Models
- GANs vs VAEs
- Introduction and GAN theory.
INCLUDED IN THE MODULE:
Boxed solutions for NLP
RNN + LSTM
RNN + LSTM.
RNN + LSTM.
Transformers
Transformers.
Boxed solutions for NLP
- regex
- "nlp swiss knives": spacy etc.
RNN + LSTM
- LSTM motivation: working with sequences
- Embeddings
- Word2Vec.
RNN + LSTM.
- Activation functions
- Vanishing/exploding gradient
- GRU/LSTM.
RNN + LSTM.
- Types of gateways
- Skip-gram
- Selection of negative examples
- GloVe.
Transformers
- Transfer Learning
- Embeddings and Language Models
- Attention in Detail: (Masked) Self-Attention, Positional Encoding, and Layer Normalization
- Transformers Variants: BERT, GPT/GPT-2/GPT-3, DistillBERT, T5, etc.
- GPT4 Demos, working with API
- Work with other current models: Mistral, LLama, Vicuna, etc.
Transformers.
- Rise of Transformers.
INCLUDED IN THE MODULE:
Introduction
Techniques
Transformers
Applications
Prompt Hub
Models
ChatGPT / Code Llama / Flan / Gemini / Gemini Advanced / Gemini 1.5 Pro / Gemma / GPT-4 / LLaMA / Mistral 7B / Mistral Large / Mixtral / OLMo / Phi-2 / Sora / LLM Collection.
Risks & Misuses
LLM Research Findings
Additional Readings
Introduction
- LLM Settings
- Basics of Prompting
- Prompt Elements
- General Tips for Designing Prompts
- Examples of Prompts.
Techniques
- Zero-shot Prompting / Few-shot Prompting / Chain-of-Thought Prompting / Generate Knowledge Prompting
- Self-Consistency / Prompt Chaining / Tree of Thoughts
- Retrieval Augmented Generation
- Automatic Reasoning and Tool-use / Automatic Prompt EngineerActive-PromptDirectional Stimulus PromptingProgram-Aided Language ModelsReActReflexionMultimodal CoTGraph Prompting.
Transformers
- Transfer Learning
- Embeddings and Language Models
- Attention in Detail: (Masked) Self-Attention, Positional Encoding, and Layer Normalization
- Transformers Variants: BERT, GPT/GPT-2/GPT-3, DistillBERT, T5, etc.
- GPT4 Demos, working with API
- Work with other current models: Mistral, LLama, Vicuna, etc.
Applications
- Function Calling
- Generating Data
- Generating Synthetic Dataset for RAG
- Tackling Generated Datasets Diversity
- Generating Code
- Graduate Job Classification Case Study
- Prompt Function.
Prompt Hub
- Classification
- Coding
- Creativity
- Evaluation
- Information Extraction
- Image Generation
- Mathematics
- Question Answering
- Reasoning
- Text Summarization
- Truthfulness
- Adversarial Prompting.
Models
ChatGPT / Code Llama / Flan / Gemini / Gemini Advanced / Gemini 1.5 Pro / Gemma / GPT-4 / LLaMA / Mistral 7B / Mistral Large / Mixtral / OLMo / Phi-2 / Sora / LLM Collection.
Risks & Misuses
- Adversarial Prompting
- Factuality
- Biases.
LLM Research Findings
- LLM Agents
- RAG for LLMs
- Trustworthiness in LLMs
- LLM Tokenization
- What is Groq?
Additional Readings
- Papers
- Tools
- Notebooks
- Datasets.
CHOOSE YOUR PLAN
BEGINNER
50 $ 70 $
- Access to pre-recorded lessons
- 🧐 Homework review
- 💻 Self-paced learning with video tutorials
- 📚 Access to course materials
IMMERSIVE LEARNING
150 $ 210 $
- Live online lessons via Zoom
- Homework feedback from instructors
- 😍 Direct interaction with instructors
- 📌 Project defense with feedback
- 🔧 Access to practical coding sessions
- 📈 Hands-on experience with AI tools
EXPERT LEVEL 🔥🔥🔥
250 $ 350 $
- Live online lessons with Zoom sessions
- Personalized feedback on assignments
- 😍 Direct communication with instructors
- 📌 Project defense with detailed review
- 🤩 One-on-one career coaching session
- 🎓 Access to exclusive industry insights
- 🖥️ Advanced projects in Machine Learning & Generative AI
Q&A
Q&A
The course is suitable for you if you want to learn the basics of working with artificial intelligence, are an aspiring or practicing IT specialist with a desire to work on AI projects, or are
an entrepreneur developing an AI product.
The course will help you:
- Gain sufficient technical expertise to work with machine learning and artificial intelligence
- Learn to manage a team of engineers and data analysts
- Plan to build, manage and scale products with machine learning and AI.
- Gain sufficient technical expertise to work with machine learning and artificial intelligence
- Learn to manage a team of engineers and data analysts
- Plan to build, manage and scale products with machine learning and AI.
Examine the data and identify factors affecting bicycle demand:
- Conduct data preprocessing.
- Conduct an EDA.
- Prepare the data to build the model.
- Implement a basic regression model for predicting the number of bicycles rented.
Some of the key differentiators of CodeMind Academy courses are:
- Live classes in an online format. This will give you the opportunity to interact directly with faculty members who are experienced professionals from large companies.
- Up-to-date class program. The course program is aligned with the latest trends in machine learning and artificial intelligence. This will enable you to learn relevant skills and gain knowledge of the latest tools to become a highly skilled professional.
- Extensive training experience. CodeMind Academy team has launched around 30 programs and curricula in the field of Artificial Intelligence and Technology. We took into account successful practices, which allowed us to develop an effective course
- Guest meetings with practitioners. The program includes guest meetings with professionals from the US, UK, European and Asian markets. This will help you to learn from successful experience and get personal advice from professionals.
There is no knowledge requirement about machine learning and AI for the course.
That said, we recommend taking a free consultation and entrance test to assess your skills and see which ones to improve. In addition to the course, you will also get free access to the Python Intensive for Generative AI if you want to refresh your existing knowledge or prepare for the start of the course.
That said, we recommend taking a free consultation and entrance test to assess your skills and see which ones to improve. In addition to the course, you will also get free access to the Python Intensive for Generative AI if you want to refresh your existing knowledge or prepare for the start of the course.
Classes will be held twice a week on Mondays/Tuesdays and Fridays at 19:00 Moscow time in Zoom. Classes include lab work, programming in pairs with classmates, development of projects
according to a ready-made technical assignment or for your task, as well as defense of projects in front of teachers. All classes will be available as recordings.
The course consists of 27 lessons (54 academic hours) and will last 4 months (14 weeks).
CodeMind Academy School has a state license and upon completion of the course you will receive a diploma of additional education.
Yes, you can get back up to 13% of the cost of the program as a tax deduction.