Data Science for Chemists

IPL Summer School, CPE Lyon

5. Machine Learning

John Samuel
CPE Lyon

Year: 2023-2024
Email: john.samuel@cpe.fr

Goals

1. Artifical Neural Networks
2. Deep Learning
3. Reinforcement Learning
4. Data Licences, Ethics and Privacy

• Inspired by biological neural networks
• Collection of connected nodes called artificial neurons.
• Artificial neurons can transmit signal from one to another (like in a synapse).
• Signal between artificial neurons is a real number
• The output of a neuron is the sum of weighted inputs.

Perceptron

• Algorithm for supervised learning of binary classifiers
• Binary classifier

Perceptron: Implementation

• Let y = f(z) be output of perceptron for an input vector z
• Let y = f(z) be output of perceptron for an input vector z
• Let N be the number of training examples
• Let X be the input feature space
• Let {(x₁, d₁),...,(x_N, d_N)} be the N training examples, where
  • x_i is the feature vector of i^th training example.
  • d_i is the desired output value.
  • x_j,i be the i^th feature of j^th training example.
  • x_j,0 = 1

Perceptron: Formal definition

• Weights are represented in the following manner:
  • w_i is the i^th value of weight vector.
  • w_i(t) is the i^th value of weight vector at a given time t.

Perceptron: Steps

1. Initialize weights and threshold
2. For each example (x_j, d_j) in training set
   • Calculate the weight: y_j(t)=f[w(t).x_j]
   • Update the weights: w_i(t + 1) = w_i(t) + (d_j-y_j(t))x_j,i
3. Repeat step 2 until the iteration error 1/s (Σ |d_j - y_j(t)|) is less than user-specified threshold.

Backpropagation

• Backward propagation of errors
• Adjust the weight of neurons by calculating the gradient of the loss function
• Error is calculated and propagated back to the network layers

Deep neural networks

• Multiple hidden layers between the input and output layers

Applications

• Computer vision
• Speech recognition
• Drug design
• Natural language processing
• Machine translation

Convolutional deep neural networks

• Analysis of images
• Inspired by neurons in the virtual cortex
• Network learns the filters

• Inspired by behaviourist psychology
• Actions to be taken in order to maximize the cumulative award.

• Data usage licences
• Confidentiality and Privacy
• Ethics

Online resources

• Artificial Neural Network
• Perceptron

Colors

• Color Tool - Material Design

Images

• Wikimedia Commons