Teaching

Here is a list of some courses I’ve delivered:

• A long intro course to Machine Learning delivered at the National Statistical Institute: David Ríos Insua, Victor Gallego Alcalá, Roi Naveiro Flores and Alberto Torres Barrán.
• An intro short course to Bayesian methods in AI: David Ríos Insua and Roi Naveiro Flores
• An intro short course to Machine Learning: David Ríos Insua and Roi Naveiro Flores.

 

A course in Risk Analysis delivered at Aalto (and Shanghai Science Tech)

• Lecture 0: Intro
• Lecture 1: Problem Structuring. IDs
• Lecture 2: Modeling beliefs
• Lecture 3: Modelling preferences under uncertainty. Utilities
• Lecture 4: Decision analytic computations
• Lecture 5: Framework for risk analysis, with applications
• Lecture 6: Noncooperative games
• Lecture 7: Adversarial risk analysis

A course in Introduction to Machine Learning at ICMAT 

• Course 0: Organization
• Course 1: General Intro
• Course 2: Regression models
  
• Course 3: Classification. Basic concepts
• Course 4: Tree based algorithms for Regression and Classification
  
• Course 5: Probabililisic Graphical Models
• Course 6. Support Vector Machines
• Course 7: Neural nets and deep learning
• Course 8: Unsupervised learning
• Course 9: Reinforcement learning

• Course 0: Motivation
• Course 1: 1. Key concepts through basic models
• Course 2: 1.2.Means – Normal models
• Course 3: 1.3. Dynamic Linear Models
• Course 4: 2. Overview of Bayesian computational methods
• Course 5: 3.1. Intro to MCMC From Gibbs to Hamilton
• Course 6: 3.2. To MCMC From Gibbs to Hamilton
• Course 7: 4. Computational method for Bayesian decision analysis
• Course 8: 5. Large scale Bayesian Inference
• Course 9: 6. Bayesian NNs/DL

 

Reading Group: Probabilistic Machine Learning: Advanced Topic. By Kevin Patrick Murphy. MIT Press, 2023.
Organizers: David Ríos Insua, Pablo García Arce, Daniel Corrales Alonso, Mario Chacón Falcón (ICMAT-CSIC)

• Session 1: Intro & Probabilistic Inference Overview (Ch. 1-7). David Ríos Insua
• Session 2: Variational Inference (Ch. 10). Pablo G. Arce
• Session 3: Monte Carlo & Hamiltonian Methods (Ch. 12). Miguel Santos Pascual
• Session 4: Sequential Monte Carlo (Ch. 13). Mario Chacón Falcón
• Session 5: VI + MCMC (Special Session). Max Hird
• Session 6: Bayesian Neural Networks (Ch. 17). Daniel Corrales Alonso
• Session 7: Beyond the i.i.d. assumption: Distributional shifts (Ch. 19). Carlos G. Meixide
• Session 8: Gaussian Processes (Ch. 18). Simón Rodríguez Santana
• Session 9: Generative models: Variational Autoencoders and Autoregressive Models (Ch. 20-22). David Ríos Insua
• Session 10: Normalizing Flows (Ch. 23). Roi Naveiro
• Session 11: Energy-Based Models (Ch. 24). Víctor Gallego
• Session 12: Diffusion Models (Ch. 25). Alberto Suárez
• Session 13: State-Space Models (Ch.29). Bruno Flores
• Session 14: Decisions (Ch. 34). José Manuel Camacho Rodríguez
• Session 15: Causality (Ch. 36). Fátima Sánchez Cabo

 

Reading Group: Intro to LLM: Coordinators: Roi Naveiro (CUNEF University) and David Rios Insua (ICMAT-CSIC)

• Session 1: Intro to Deep Learning focusing on LSTMs and GRUs 
• Session 2: Formal Algorithms for Transformers
• Session 3: Generative Pre-Training (GPT)
• Session 4: Reinforcement Learning from Human Feedback (RLHF)

 

Reading Group in Causal Inference and Machine Learning at ICMAT with Roi Naveiro, Jose Manuel Camacho, Juan Poyatos, Ramón Díaz-Uriarte, Alex Pozas, Simón Rodriguez Santana, Fátima Sánchez Cabo

• Session 1: Book: What If? Miguel Hernán. Chapter 1: A definition of causal effects
• Session 2: Book: What If? Miguel Hernán. Chapter 2: Randomized experiments
• Session 3: Book: What If? Miguel Hernán. Chapter 3: Observational Studies
• Session 4: Book: What If? Miguel Hernán. Chapter 4: Effect modification
• Session 5: Book: What If? Miguel Hernán. Chapter 5: Interaction
• Session 6: Book: What If? Miguel Hernán. Chapter 6: Graphical Representation of causal effects
• Session 7: Book: What If? Miguel Hernán. Chapter 8: Selection Bias
• Session 8: Book: What If? Miguel Hernán. Chapter 7: Confounding
• Session 9: Quantum information tools that deal with causality. The session is based on: [1], [2], [3]
• Session 10: Book: What If? Miguel Hernán. Chapter 9: Measurement bias
• Session 11: Book: What If? Miguel Hernán. Chapter 10: Random Variability
• Session 12: Casually (and causally?) playing with a large health database.
• Session 13: Mendelian Randomization: Using genetic variability as an instrumental variable in Causal Inference. Lecturer: Fátima Sánchez Cabo