You are here

Unsupervised Learning and non-Parametric methods

External Lecturer: 
Alessandro Laio
Alex Rodriguez
Course Type: 
PhD Course
Academic Year: 
January - February
38 h

10 x 2 hours + 6 x 3 hrs Labs = 38 hours (6 weeks: 11/01-19/02)

1. Introduction: choosing the features and the metric.
2. Lab 1
3. Dimensional reduction and manifold learning
    i. Linear methods: principal component analysis and multidimensional scaling
    ii. Curved manifolds: ISOMAP, kernel PCA and Sketchmap
    iii. Lab 2
    iv. Diffusion Map and Stochastic Neighbor Embedding
    v. Characterizing the embedding manifold: the intrinsic dimension
    vi. Lab 3
4. Estimating the probability density
    i. Parametric density estimators
    ii. Non-parametric estimators: Histograms, Kernel density estimator and k-nearest neighbor estimator
    iii. Adaptive density estimators
    iv. Lab 4
5. Clustering
    i. Partitioning schemes: k-means, k-medoids and k-centers.
    ii. hierarchical and spectral clustering
    iii. Lab 5
    iv. Density-based clustering
    v. Clustering techniques exploiting kinetic information
    vi. Lab 6

Please, notice that this is a course belonging to Data Science Excellence Department programme. MAMA PhD students can plan 33% of their credits (i.e. 50 hrs) from this programme.

Next Lectures: 

Sign in