Data scientist with reinforcement learning

Fundamentals of Data Engineering and Data Architectures (approx. 2 days)

Tasks and Application Areas in Data Engineering

Fundamentals of Modern Data Architectures and Data Platforms

Fundamentals and Comparison of Cloud-Based Data Architectures

OLTP and OLAP

Processing Structured, Semi-Structured, and Unstructured Data

Requirements Management and Data Modeling (approx. 1 day)

Requirements for Data and Data Sources

Conceptual Data Modeling

Modeling Data Structures and Data Flows

Data Contracts, Data Governance, and Data Lineage

Databases and Data Processing (approx. 1 day)

Database System Architecture

SQL vs. NoSQL vs. NewSQL: A Comparison

Relational Data Modeling

SQL for Querying and Processing Data

Data Warehousing and Modern Data Architectures (approx. 2 days)

Fundamentals and Architecture of Data Warehouse Systems

Dimensional Data Modeling and Schema Design

Facts, Dimensions, and Historical Data Storage

Comparing Data Warehouses, Data Lakes, and Lakehouses

Introduction to dbt

Practical Implementation of Data Models

Data Integration and Data Pipelines (approx. 5 days)

Data Integration and Data Provisioning

ETL and ELT Processes

Design and Orchestration of Data Pipelines

Data Cleaning, Validation, and Data Quality

Batch and Streaming Processing

Practical Application of Airflow and Spark

Programming and Automated Data Processing (approx. 4 days)

Python Environments and Package Management

Data Processing and Transformation with Python

Automation of data processing workflows

Connecting to Different Data Sources and Interfaces

Development of data engineering workflows

Containerization with Docker

AI-powered support in the development and data processing process

Project work (approx. 5 days)

To consolidate the content learned

Presentation of the project results

Introduction to Data Analytics (approx. 1 day)

Tasks, Areas of Application, and Typical Use Cases of Data Analytics

Data Analytics Processes and CRISP-DM

The Role of the Data Analyst and How It Differs from Data Science, Data Engineering, and Business Intelligence

Classification of Artificial Intelligence and Machine Learning

Python for Data Analytics (approx. 1 day)

Working with Jupyter Notebooks

NumPy and Pandas for Data Analysis

Importing and Exporting Data from Various Sources, e.g., Excel and CSV

Selecting and transforming data

Data Preparation and Exploratory Data Analysis (approx. 3 days)

Data cleaning and transformation with Pandas

Handling missing, incorrect, and outlier data

Merging, grouping, and aggregating data

Exploratory data analysis and deriving initial insights

Fundamentals of statistical analysis

SQL and Data Management (approx. 2 days)

Relational Databases and Data Models

Data Queries and Processing with SQL

Joins, aggregations, and analytical queries

Interaction Between SQL and Python in Data Analysis

Data Quality and Data Protection

Connecting to and Using Different Data Sources and Database Systems

Data Visualization and Data Storytelling (approx. 2 days)

Visualization with Matplotlib, Seaborn, and Plotly

Selecting Appropriate Charts and Visualization Methods

Preparing and interpreting analysis results

Data Storytelling and Communication Tailored to the Audience

Power BI and Interactive Dashboards (approx. 2 days)

Data Import and Preparation with Power Query

Data modeling and relationships

Basics of DAX and metrics

Creating interactive reports and dashboards

Visualization and Communication of Business KPIs

Overview of Alternative Business Intelligence and Visualization Solutions

AI-powered data analysis (approx. 2 days)

Use of generative AI in the data analytics workflow

AI-powered data preparation, analysis, and visualization

Support with Python and SQL queries

Prompting for data analysis and interpretation of results

Critical evaluation of data sources and detection of synthetic and implausible data

Quality assurance, data protection, and responsible use of AI

Business Analytics and Statistical Fundamentals (approx. 2 days)

Key Metrics and KPIs in a Business Context

Descriptive Statistics and Statistical Relationships

Correlation, Causality, and Simple Regression Analyses

Fundamentals of confidence intervals and hypothesis testing

Interpreting Analysis Results and Deriving Data-Driven Recommendations

Project work (approx. 5 days)

To consolidate the content learned

Presentation of the project results

Introduction to Machine Learning and the ML Workflow (approx. 3 days)

Fundamentals, Applications, and Types of Machine Learning

Classification of Modern AI and Machine Learning Methods

Machine Learning Workflow with scikit-learn

Training, validation, and test data

Underfitting, overfitting, and model evaluation

Cross-Validation

Data Preparation and Feature Engineering (approx. 2 days)

Data Quality and Preprocessing

Scaling, Normalization, and Encoding

Feature Engineering and Feature Selection

Handling Missing Values, Outliers, and Imbalanced Datasets

Supervised Learning – Regression and Classification (approx. 5 days)

Regression and Classification Methods

Linear and Regularized Models

Distance-, probability-, and tree-based methods

Ensemble and Boosting Methods

Model Comparison and Hyperparameter Optimization

Evaluation Metrics and Probabilistic Predictions

Feature importance and explainable AI 

Unsupervised Learning (approx. 2 days)

Dimension Reduction and Feature Extraction

Clustering Methods and Evaluation of Clustering Results

Anomaly Detection

Time Series Analysis and Forecasting (approx. 1 day)

Fundamentals and Analysis of Time Series Data

Forecasting Methods and Machine Learning for Time Series

Neural Networks and Modern AI (approx. 2 days)

Fundamentals of Neural Networks and Deep Learning

Machine Learning for Text Data and Natural Language Processing

Foundation Models, Large Language Models, and Generative AI

Deployment of ML models (approx. 1 day)

Storage and deployment of trained models

Development of simple ML applications

Interfaces and platforms for machine learning models

Project work (approx. 4 days)

To consolidate the content learned

Presentation of the project results

Introduction to Deep Learning (approx. 1 day)

Deep learning as a type of machine learning

Fundamentals of neural networks (approx. 4 days)

Multilayer perceptrons

Calculation of neural networks

Optimization of model parameters, backpropagation

Deep learning libraries

Regression vs. classification

Typical loss and activation functions

Evaluating the model prediction with metrics

Regression and classification metrics

Learning curves, overfitting and regularization

Hyperparameter optimization

L1/12 regularization

Dropout

Early stopping

Stochastic gradient descent (SGD)

Momentum, Adam Optimizer

Optimization of the learning rate

Dynamic learning rate adjustment

Reduce Learning Rate on Plateau

Learning rate optimization with the TensorBoard

Controlling the fit process with callbacks

Saving and loading models

Convolutional Neural Network (CNN) (approx. 2 days)

Image classification

Convolutional layers, pooling layers

Reshaping layers, flattening, global-average pooling

CNN architectures ImageNet-Competition

Deep neural networks, vanishing gradients, skip connections, batch normalization

Transfer learning (approx. 1 day)

Adapting and combining models

Unsupervised pre-training

Image data augmentation, explainable AI

Data loader

Regional CNN (approx. 1 day)

Object localization

Semantic segmentation

Regression problems

Branched neural networks

YOLO architecture

U-Net models

Methods of creative image generation (approx. 1 day)

Generative Adversarial Networks (GAN)

Deepfakes

Diffusion models

Superresolution

Supplementing image areas

Apply foundation models from Hugging Face

Multimodal models

LoRA-Fine-Tuning

Application areas of generative models

Legal restrictions

Recurrent neural networks (approx. 2 days)

Sequence analysis

Recurrent layers

Backpropagation through time (BPTT)

Analysis of time series

Exploding and vanishing gradient problems

LSTM (Long Short-Term Memory)

GRU (Gated Recurrent Unit)

Deep RNN

Deep LSTM

Text processing using neural networks (approx. 2 days)

Text preprocessing

Embedding layers

Text classification

Sentiment analysis

Natural Language Processing (NLP)

Translations

Text generation

Sequence-to-sequence method, encoder-decoder architecture

Encoder-only and decoder-only models

Local application of large language models

Language models (approx. 1 day)

Transfomer architecture

Attention and Multihead-Attention

Positional Encodings

Fine-tuning large language models

Prompting

Text generation pipelines

Summarization

Chatbots

Retrieval Augmented Generation

AI agents

Deep reinforcement learning (approx. 1 day)

Control of dynamic systems

Agent systems

Training through rewards

Policy Gradients

Deep Q-learning

Bayesian neural networks (approx. 1 day)

Uncertainties in neural networks

Statistical evaluation of forecasts

Confidence, standard deviation

Unbalanced data

Sampling methods

Project work (approx. 3 days)

To consolidate the content learned

Presentation of the project results

Introduction to reinforcement learning (approx. 1 day)

Definition and basic concepts

Differences to other learning methods

Areas of application and examples

Markov Decision Processes (MDPs) (approx. 2 days)

Definition and properties of MDPs

Value functions and policy

Bellman equations

Dynamic Programming Approach

Q-Learning (approx. 2 days)

Definition and algorithm

Exploration vs. exploitation

Convergence and optimization properties

Applications in games, robotics and other areas

Deep reinforcement learning (approx. 3 days)

Deep Q-Learning

Deep Deterministic Policy Gradients (DDPG)

Actor Critical Methods

Policy Gradient Methods

Advanced topics (approx. 4 days)

Model-Based Reinforcement Learning

Multi-Agent Reinforcement Learning

Inverse Reinforcement Learning

Meta Reinforcement Learning

Practical applications (approx. 3 days)

Implementation of reinforcement learning algorithms

Application to selected problems and case studies

Evaluation and tuning of the algorithms

Summary and outlook (approx. 2 days)

Summary of the most important concepts and results

Challenges and future developments in reinforcement learning

Project work (approx. 3 days)

To consolidate the content learned

Presentation of the project results