Search Colleges, Exams, Schools & more
Interested in this College?
Get updates on Eligibility, Admission, Placements Fees Structure
CompareGet updates on Eligibility, Admission, Placements Fees Structure
Quick Facts
| Medium Of Instructions | Mode Of Learning | Mode Of Delivery |
|---|---|---|
| English | Self Study, Virtual Classroom | Video and Text Based |
Courses and Certificate Fees
| Certificate Availability | Certificate Providing Authority |
|---|---|
| yes | Imperial College Business School, London |
The Syllabus
- Meet the faculty and get an overview of the programme, including certificate requirements, the learning platform and support resources. Receive an introduction to Python and learn about the capstone competition
- Become familiar with the fundamental components of and approaches to machine learning problems and ways to classify problems along the major dividing lines of the ML landscape
- Understand how to differentiate ML from statistics
- Examine real-world applications of machine learning across a variety of industries
- Learn how to calculate absolute, conditional and total probabilities and understand how to classify independent versus dependent events
- Discover how to run simulations using random number generating libraries in Python and Numpy
- Explore the difference between discrete and continuous random variables and learn how to compute probabilities and values related to binomial distribution
- Understand how to conduct maximum likelihood estimations using existing data
- Understand how to detect outliers and calculate regression or correlation coefficients fora data set
- Examine the potential consequences of eliminating outliers from a data set
- Explore the validity of statements about data sets and relationships among data
- Learn how to compute confidence interval
- Recognise when it is feasible to draw meaningful conclusions from data
- Learn how to calculate the probability of selecting the correct model from a set of sample data
- Examine key components of generalisation bounds
- Learn how to estimate the fit of a selected model on a new data set
- Learn how to use performance measures to evaluate regression problems with a numerical output variable
- Understand how to apply a confusion matrix to evaluate classification problems with a categorical output variable
- Examine challenges that can be addressed by machine learning competitions across a variety of industries
- Explore effective applications of oversampling to a machine learning problem
- Understand how to apply oversampling to classification problems
- Learn how to estimate the performance of a given predictor using the k-fold cross-validation algorithm
- Learn how to calculate distance functions for k-nearest neighbour methods and understand how to apply normalisation methods to scale data sets
- Learn how to use validation and test sets to predict and select the value for regression and classification problems
- Explore real-life applications of k-nearest neighbour methods that take into consideration the advantages and shortcomings of such methods
- Discover how a decision tree makes predictions and understand the difference between howto measure purity in categorical models with entropy and the Gini Index
- Explore how a computer constructs a decision tree and examine strategies for pruning a classification tree
- Recognize the differences between constructing regression and classification trees
- Learn how to select the most appropriate tree depth for making a prediction
- Become familiar with the concepts of interpretability, fairness and non-discrimination
- Explore the functionality of the k-nearest neighbour and decision tree methods
- Understand key components of Bayes’ theorem and learn how to use Bayes' theorem to calculate conditional probabilities
- Recognize when it is better to predict a probability instead of an actual value
- Explore real-life applications of the Naïve Bayes theorem
- Understand the use of surrogate models as well as their applications and pitfalls
- Explore machine learning algorithm parameters and the most common surrogate methodsused for tuning
- Examine trade-offs between exploration and exploitation in Bayesian optimisation
- Discover when continued parameter tuning is no longer worthwhile
- Compare logistic and linear regression to understand the categorical output and best use cases of logistic regression for binary classification
- Using a real-life data set, apply the maximum likelihood method to fit a logistic regression
- Select an approach for the final code base of your capstone project that aligns with your professional goals
- Understand the concepts of linear separation through hyperplanes
- Learn about hard-margin and soft-margin support vector machines
- Gain knowledge of the kernel trick and discover how support-vector machines can be applied to classification problems with more than one outcome
- Explore similarity measures between samples and clusters
- Learn about hierarchical and k-means clustering
- Discover common concerns that arise with cluster analysis
- Examine real-life applications of clustering
- Delve deeper into calculating the direction and location of principal components
- Learn how to determine the number of principal components to use
- Become familiar with the history of deep learning and discover the five building blocks of deep learning
- Understand the function approximation technique
- Learn how the backpropagation algorithm is used to train a neural network using thechain rule technique
- Understand the backward pass component of backpropagation
- Discover how to optimise parameters using stochastic gradient descent
- Understand how to fine-tune machine learning models using a hyperparameter search
- Discover how to use regularisation techniques to make adjustments that will improve model performance
- Receive an introduction to using PyTorch
- Explore examples of neural networks gone wrong
- Learn to interpret the output of a neural network
- Understand how to design machine learning models with interpretability in mind
- Receive an introduction to convolutional neural networks
- See the role convolutions play in computer vision
- Examine the building blocks of the LeNet-5 architecture
- Learn how to build convolutional neural networks using PyTorch
- Understand the differences between model-based and model-free approaches
- Learn how to combine reinforcement and supervised learning
- Discover how to build systems to create reinforcement learning experiences
- Discover how to construct appropriate surrogate models
- Understand how to balance exploration and exploitation
- Learn how to manage different types of variables
Instructors
Back to top