:::danger Initial drafts, could be outdated. Scroll below for actual maps ::: ```markmap # MIMER Training Program ## Core AI Foundations ### Machine Learning - Foundations of Machine Learning: Theory to Practice - Applied Machine Learning for Real-World Data - Scalable & Efficient Machine Learning on HPC Systems ### Deep Learning - Deep Learning Fundamentals - Advanced Architectures for Vision and Sequence Modeling - Scalable Deep Learning on GPUs and HPC Systems - Optimization, Regularization, and Model Compression ### Generative AI - Foundations of Generative AI - Large Language Models (LLMs): Fine-Tuning and Adaptation - Multimodal & Diffusion Models for Content Generation - Scalable & Trustworthy Generative AI Systems - Generative coding (for beginners/advanded) - Local LLM, Ollama and local agents --- ## AI Infrastructure & Systems ### AI Systems & Engineering - Distributed AI - MLOps & LLMOps - AI Infrastructure & HPC Integration - Model Deployment & Serving - AI Optimization & Acceleration - Edge AI Systems ### Data Engineering for AI - Data Governance - Data Quality & Curation - Big Data Architectures - Data Security & Privacy - Synthetic Data Systems ### AI Security - Adversarial Machine Learning - Model Attacks & Defenses - Secure Model Deployment - AI Red Teaming - Robustness Testing - Secure AI agents --- ## Advanced AI Paradigms ### Reinforcement Learning - Reinforcement Learning - Multi-Agent Systems - Control Systems - Optimization & Planning - Autonomous Systems AI ### AI for Scientific Computing - Physics-Informed AI - Surrogate Modeling - AI-Accelerated Simulation - Hybrid AI + Numerical Methods - Scientific Foundation Models ### Computer Vision & Perception - Image Analysis - Video Understanding - 3D Vision - Remote Sensing AI - Medical Imaging AI ### Natural Language Processing - Text Analytics - Information Retrieval - Knowledge Graphs - Speech & Audio AI - Multilingual AI Systems ### Multimodal AI - Vision-Language Models - Audio-Visual AI - Cross-Modal Retrieval - Embedding Systems --- ## Responsible & Sovereign AI ### Trustworthy & Responsible AI - Explainable AI (XAI) - Fairness & Bias Mitigation - AI Safety & Robustness - Privacy-Preserving AI - AI Governance & Compliance - Risk Assessment Frameworks ### AI Security & Sovereignty ### AI Governance & Policy Compliance ``` ## This tree older (see below for treee in D3.1): ```markmap ## Domain AI Applications ### AI for Industry & Society - AI Strategy & Digital Transformation for Enterprises - AI Adoption Roadmaps for SMEs - ROI Assessment and Business Case Development for AI - AI for Risk Management and Compliance - Human–AI Collaboration in the Workplace - Ethical and Trustworthy AI in Industrial Deployment - AI Governance & EU AI Act Readiness - AI for decision makers ### AI for Manufacturing - AI for process modelling - Reinforcement learning for process control ### AI for Life Science - AI for Genomics and Multi-Omics Data Analysis - Deep Learning for Medical Imaging - AI in Drug Discovery and Molecular Design - Protein Structure Prediction with AI - AI for Clinical Decision Support Systems - AI for Epidemiological Modeling - Digital Twins in Personalized Medicine - Responsible AI in Healthcare Applications ### AI for Materials Science - AI for Materials Discovery and Design - Machine Learning for Computational Chemistry - Surrogate Modeling for Simulation Acceleration - Physics-Informed Neural Networks (PINNs) - AI for Microstructure Analysis - Generative Models for Novel Material Design - AI-Enhanced Molecular Dynamics Simulations - Autonomous Materials Experimentation ### AI for Autonomous Systems - Core Perception & Sensing - Decision-Making & Control - Reinforcement learning for robotic control ### AI for the Gaming Industry - Game AI Fundamentals - Advanced AI for Gameplay - Generative AI for Games - AI for Player Experience & Analytics - Technical & Scalable AI in Gaming - AI for ray tracing and neural rendering ``` ```markmap # AI Skill Tree ``` ```markmap ## Foundations of AI ### Mathematical Foundations - Linear Algebra - Calculus - Probability & Statistics - Optimization Theory ### Programming Fundamentals - Python - Data Structures - Algorithms ### Data Handling - Data Cleaning - Feature Engineering - Exploratory Data Analysis (EDA) --- ``` ```markmap ## Machine Learning ### Basic ML Frameworks - scikit-learn ### Supervised Learning - Regression - Classification - Ensemble Methods ### Unsupervised Learning - Clustering - Dimensionality Reduction ### Model Evaluation - Cross-Validation - Evaluation Metrics ### Practical ML - Overfitting / Underfitting - Regularization - Data Leakage Prevention --- ``` ```markmap ## Deep Learning ### Core Concepts - Neural Network Architectures - Forward / Backward Propagation - Activation Functions - Loss Functions ### Deep Learning Frameworks - PyTorch - TensorFlow / Keras ### Training Techniques - Optimization Algorithms - Dropout & Batch Normalization - Learning Rate Schedulers ### Computer Vision - CNNs - Image Segmentation - Object Detection ### Natural Language Processing - RNNs, LSTMs, GRUs - Transformers - Pretraining & Fine-Tuning --- ``` ```markmap ## Generative AI ### Generative Models - Variational Autoencoders (VAEs) - Generative Adversarial Networks (GANs) ### Large Language Models (LLMs) - Architecture & Attention Mechanisms - Prompt Engineering - Retrieval-Augmented Generation (RAG) - Fine-Tuning & LoRA - Agents ### Multimodal Models - Vision-Language Models - Diffusion Models - Speech-to-Text / Text-to-Speech --- ``` ```markmap ## Reinforcement Learning ### Foundations - Markov Decision Processes - Policies, Rewards & Value Functions ### Core Algorithms - Q-Learning - SARSA - Deep Q-Networks (DQN) ### Advanced RL - Policy Gradient Methods - Actor-Critic Methods - Multi-Agent RL --- ``` ```markmap ## MLOps & AI Engineering ### Data Pipelines - Data Versioning - Streaming & Batch Processing ### Model Deployment - REST APIs - Containerization (Docker) - Cloud Platforms (Azure, AWS, GCP) ### Model Monitoring - Drift Detection - Model Retraining Cycles ### CI/CD for ML - Automated Training Pipelines - Experiment Tracking (MLflow, Weights & Biases) --- ``` ```markmap ## AI Safety, Ethics & Governance ### Bias, Fairness and Ethics - Dataset Bias - Algorithmic Fairness Techniques - AI Ethics ### Explainability - SHAP - LIME - Interpretability Tools ### Privacy - Privacy-Preserving AI - Differential Privacy - Secure Multiparty Computation - Homomorphic Encryption - Federated Learning - Privacy in LLMs and generative AI ### Security & Robustness - Adversarial Attacks - Model Hardening - Formal verification - Counter example driven optimization ### Cybersecurity - Secure Coding for AI engineering - Secure AI agents and LLMs - Secure AI workflows ### Policies & Compliance - Responsible AI - Governance Frameworks - Regulatory Standards (EU AI Act, etc.) - Privacy regulation and GDPR - Cybersecurity standards and guidelines for AI ``` ## These trees will be in D3.1: ```mermaid mindmap root((Curriculum)) Foundational AI Generative AI Applied And Advanced AI Focus Areas Industry & Society Life Sciences Materials Science Autonomous Systems Gaming ``` ```markmap ## Foundational AI ### Machine Learning - Practical Machine Learning - Machine Learning for Real-World Tabular Data - Scalable & Efficient Machine Learning on HPC Systems - Natural Language Processing ### Deep Learning - Introduction to Deep Learning - Convolutional Neural Networks - Recurrent Neural Networks - Transfer Learning and Fine Tuning - Group Relative Policy Optimization (GRPO) - Scalable and Resource-efficient Deep Learning ### AI Development Life Cycle - Introduction to MLOps - Introduction to LLMOps - Data preparation and representation - Hyperparameter Tuning - Inference-optimized AI ``` ```markmap ## Generative AI ### Tasks and Paradigms - Large Language Models (LLMs) - Vision Language Models - Audio Models for STT and TTS - Diffusion Models ### Agentic AI - Responsible Use of Generative AI in Assisted Coding - Retrieval Augmented Generation - Agentic AI 101 - Model Context Protocol and AI Skills - Sovereign LLM and agents ``` ```markmap ## Applied and Advanced AI ### Trustworthy AI - Ethical and Trustworthy AI self-assessment guide - Federated Machine Learning - Privacy-preserving AI - Synthetic data generation - Cybersecurity for AI systems ### AI for Science - Physical AI and Surrogate Modeling - Foundational models - Hybrid AI + Numerical Methods ### Computer Vision - OCR and handwritten text recognition - Segmentation and object detection ``` ## This tree ~~is~~ will be also in D3.1: ```markmap ## Focus Areas ### Autonomous Systems - Core Perception & Sensing - Decision-Making & Control - Reinforcement learning for robotic control - AI agents in cyberphysical systems - Classical control and AI ### Gaming - Game AI Fundamentals - AI Moderation for Safe Game environments - Game Jam and Hackathons - AI for eductional games - Game engines and AI for ray tracing and rendering ``` ```markmap ## Focus Areas ### Industry & Society - Introduction to AI for decision makers - Human–AI Collaboration in the Workplace - Ethical and Trustworthy AI for decision makers - Digital competence, security and best practices - AI for change: opportunities and risks ``` ```markmap ## Focus Areas ### Life Sciences - Generative AI for life sciences - Explainable AI for life sciences - Protein Language Models for Structure Prediction - Vision language models for Medical Imaging - Reconstructing Proteins and Molecules from Microscopy - Semantic, instance and panoptic segmentation - Feature extraction in Genomic data - Graph Neural Networks and Transformers in Life Sciences ### Materials Science - Data-driven methods for X-ray & CT - Fingerprinting of Industrial Objects - Deep Learning and Computer Vision for Defect Detection - Designing Machine Learning Potentials ``` --- --- # AI Learning Personas # Manager/Decision maker - Manager in company (general) - Manager in company (advanced AI use) - Decision maker public administration - Politician # Layperson - Citizen - Employee private or public sector (non-technical, uses AI at work or needs awareness) - Journalist (including content creator, editor etc.) - Lawyer # Student - University student STEM field - University student non-STEM field - University student quantitative field (non-STEM) - University student non-quantitative field - High school student and below # Artist - Sound (musician, sound artist) - Visual and Physical (actor, painter, illustrator, sculptor, art director, architect, digital art creator etc.) - Text (author, art director) # Technical specialist - Developer (not AI-expert) - Statistician - Engineer (not AI-expert) # AI and data specialist - AI engineer - Data scientist (include for example bio-informantics etc.) # Researcher - AI researcher - Researcher STEM subject (using AI or needs AI awareness) - Researcher non-STEM subject (using AI or needs AI awareness) - Researcher quantitative field (non-STEM) - Researcher non-quantitative field