Machine Learning

Infographics

• Supervised Learning (Linear Regression, Logistic Regression, Decision Trees, Support Vector Machines, K-Nearest Neighbors, Naive Bayes)
• Unsupervised Learning (K-Means Clustering, Principal Component Analysis)
• Reinforcement Learning (Q-Learning)

Jupyter notebooks

• Machine Learning (Overview)
• Supervised Learning (Linear Regression, Logistic Regression, Decision Trees, Support Vector Machines, K-Nearest Neighbors, Naive Bayes)
• Unsupervised Learning (K-Means Clustering, Principal Component Analysis)
• Reinforcement Learning (Q-Learning)

Deep Learning

Infographics

1. Feedforward Neural Networks (FFNN) / Multilayer Perceptrons (MLP)
2. Convolutional Neural Networks (CNN) / ConvNets
3. Recurrent Neural Networks (RNN)
4. Long Short-Term Memory (LSTM) Networks
5. Gated Recurrent Units (GRU)
6. Autoencoders (AE)
7. Generative Adversarial Networks (GANs)
8. Transformer Networks (TN)
9. Graph Neural Networks (GNNs)
10. Residual Networks (ResNets)
11. Densely Connected Convolutional Networks (DenseNets)

Jupyter notebooks

Deep Learning

• Neuron (Perceptron)
• Activation Functions
• Neural Network Architecture
• Loss Function (Cost Function)
• Optimization Algorithm (Gradient Descent)
• Backpropagation

1. Feedforward Neural Networks (FFNN) / Multilayer Perceptrons (MLP)
2. Convolutional Neural Networks (CNN) / ConvNets
3. Recurrent Neural Networks (RNN)
4. Long Short-Term Memory (LSTM) Networks
5. Gated Recurrent Units (GRU)
6. Autoencoders (AE)
7. Generative Adversarial Networks (GANs)
8. Transformer Networks (TN)
9. Graph Neural Networks (GNNs)
10. Residual Networks (ResNets)
11. Densely Connected Convolutional Networks (DenseNets)

• History - Schmidhuber

AI Agents and Scientists

Infographics

• AI Agents (§0 The loop §1 Formal setting §2 Lab: Bellman §3 Reasoning & planning §4 Tools & memory §5 Learning to act §6 Multi-agent §7 Evaluation §8 Open problems §9 Reading list)
• AI Scientists (The loop Formalism Reliability False discovery Tournaments Systems Autonomy Failure modes For your PhD Reading)

• AI Agents

References

• Luo et al. "The more you automate, the less you see: Hidden pitfalls of autonomous AI scientists"

Machine-Learned Force Fields

Infographics

• Machine-Learned Force Fields

1. J. Behler, M. Parrinello, Generalized neural-network representation of high-dimensional potential-energy surfaces, Phys. Rev. Lett. 98, 146401 (2007). doi
2. A. P. Bartók, M. C. Payne, R. Kondor, G. Csányi, Gaussian approximation potentials, Phys. Rev. Lett. 104, 136403 (2010). doi
3. A. P. Bartók, R. Kondor, G. Csányi, On representing chemical environments (SOAP), Phys. Rev. B 87, 184115 (2013). doi
4. J. S. Smith, O. Isayev, A. E. Roitberg, ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost, Chem. Sci. 8, 3192 (2017). doi
5. K. T. Schütt et al., SchNet – A deep learning architecture for molecules and materials, J. Chem. Phys. 148, 241722 (2018). doi
6. L. Zhang, J. Han, H. Wang, R. Car, W. E, Deep potential molecular dynamics, Phys. Rev. Lett. 120, 143001 (2018). doi
7. R. Drautz, Atomic cluster expansion for accurate and transferable interatomic potentials, Phys. Rev. B 99, 014104 (2019). doi
8. S. Batzner et al., E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials (NequIP), Nat. Commun. 13, 2453 (2022). doi
9. I. Batatia et al., MACE: Higher order equivariant message passing neural networks for fast and accurate force fields, NeurIPS (2022). arXiv:2206.07697
10. A. Musaelian et al., Learning local equivariant representations for large-scale atomistic dynamics (Allegro), Nat. Commun. 14, 579 (2023). doi
11. T. W. Ko, J. A. Finkler, S. Goedecker, J. Behler, A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer, Nat. Commun. 12, 398 (2021). doi
12. O. T. Unke et al., Machine learning force fields, Chem. Rev. 121, 10142 (2021). doi
13. J. Behler, Four generations of high-dimensional neural network potentials, Chem. Rev. 121, 10037 (2021). doi
14. X. Fu et al., Forces are not enough: Benchmark and critical evaluation for machine learning force fields with molecular simulations, TMLR (2023). arXiv:2210.07237
15. I. Batatia et al., A foundation model for atomistic materials chemistry (MACE-MP-0), arXiv:2401.00096

Jupyter Notebooks

• Machine-Learning Force Fields

References

• 2021 Chemical Reviews
• ANI  Espaloma  MACE

Useful links

• Claude  Colab  NotebookLM  

MetaCentrum

• Chat.ai.e-infra.cz  Matrix  Element
• Webinars  MetaCentrum  LLM and AI

History

• 2026 The Road to Modern AI
• 1965 Speculations Concerning the First Ultraintelligent Machine
• 1955 A PROPOSAL FOR THE DARTMOUTH SUMMER RESEARCH PROJECT ON ARTIFICIAL INTELLIGENCE

Gallery