TP-ComDec

Posted on Thu 23 January 2020 in posts • Tagged with python, machine learning, graph partitionning, community detection, statistical physics

TP Network and Community Detection

the goal here is to implement different methods for partitionning a graph into many clusters. We will first focus on the graph Laplacian, and then on implementing Belief Propagation.

Spectral Clustering

The first part is to construct a fonction generating random graph with blocks (stochastic block model) in the simplest manner. Then we will look at some properties using the graph spectrum.


Continue reading

TP-GMM

Posted on Thu 23 January 2020 in posts • Tagged with python, machine learning, gaussian mixture, statistical physics

TP-GMM-Annealing

Posted on Thu 23 January 2020 in posts • Tagged with python, machine learning, gaussian mixture, statistical physics, annealing

Annealing and GMM

In this program we make an annealing procedure with many centers in the GMM model. We only adjust the centers and the inverse temperature $\beta$ corresponds to the inverse of the variance of each center

TP-NN

Posted on Thu 23 January 2020 in posts • Tagged with python, machine learning, neural network, perceptron, statistical physics

TP Neural Network

the first part of this session is to implement its own version of the perceptron and get more familiar with the practice of ML.

Multiclass Perceptron

In order to make your own perceptron, you will have to implement the following

  1. Define the softmax function
  2. PeceptronOut : return the output for each of the $K$ neurons.

  3. Continue reading