Durante el mes de Enero, desde el lunes 28 hasta el viernes 1 de Febrero, se llevará a cabo el Seminario de Machine Learning. En este se cubrirán muchos de los temas actualmente relevantes en el ámbito profesional y de investigación.
Les dejamos aquí el cronograma:
| Día | Temas |
|---|---|
| 1 | 1. Introduction * History and definition * Terminology and useful concepts * Machine Learning Engineer * Tools for Machine Learning * Applications and use cases 2. Data Preparation * Data analysis and preparation * Pre-processing (cleaning, manual elimination, unbalanced data, dimension reduction, normalization, standardization) * Data exploration (histogram, box-plot, pie-chart, scatter-plot, heat map) 3. Distance-based methods * Predictive models * Distance-based models * Types of distances (Minkowski, cosine) * k-Nearest Neighbors * k-Means Clustering |
| 2 | 4. Searching-based methods * Introduction * Decission Trees * Random Forests 5. Introduction to Neural Networks * History and biological inspiration * Important definitions * Topology of Neural Networks * Structure of a Neural Network * Learning process (problems and solutions) * Example |
| 3 | 6. Support Vector Machines * History and definition * Linear SVM * Non-linear SVM * Kernel types * One-class SVM * Advantages and disadvantages 7. Naive Bayes * Bayes theorem * Naive Bayes * Example * Advantages and disadvantages 8. Probabilistic Graphical Models * Introduction * Usability * Representation * Random Markov Fields * Applications * Example |
| 4 | 9. Ensemble models * Homogeneous models (Bagging, Boosting) * Heterogeneous models (Stack, Cascade) 10. Genetic Algorithms * Introduction * Genetic Algorithms * Example 11. Evaluation * Error metrics * Sampling (bootstrap, cross validation, holdout) * Muticlass classification * Overfitting vs Underfitting |
| 5 | 10. Deep learning * Introduction to Deep Learning * Convolutional Neural Networks * Deep Learning Frameworks * Recurrent Neural Networks |
Las charlas están pensadas en cubrir los aspectos básicos de cada tema, por lo que no son necesarios conocimientos previos. Además, se mostrarán ejemplos reales sobre el uso y aplicación de las diferentes técnicas.
Lugar: Auditorio del 4to piso de Informática - UNSAAC
Hora: 5pm a 7pm
Ponentes:
- PhD. Victor Ayma (PUC-RIO)
- PhD. Student Grover Castro (IME-USP)
- PhD. Student Danny Suarez (UFRGS)
- PhD. Student Rodolfo Quispe (UNICAMP)
- M.Sc. Jhosimar Arias (UNICAMP)
- M.Sc. Berthin Torres (UNICAMP)
Nota: Ingreso libre! No es necesario traer portátiles. Se darán certificados digitales.
Material de las charlas (diapositivas + código fuente): https://drive.google.com/drive/folders/1YDxViYU8Y05hqzDv3XznAFvadw-e4vUU?usp=sharing