COM6012 Scalable Machine Learning
| Summary |
This module will focus on technologies and algorithms that can be applied to data at a very large scale (e.g. population level). From a theoretical perspective it will focus on parallelisation of algorithms and algorithmic approaches such as stochastic gradient descent. There will also be a significant practical element to the module that will focus on approaches to deploying scalable ML in practice such as SPARK, programming languages such as Python/Scala and deployment on high performance computing platforms/clusters. |
| Session |
Spring 2026/27 |
| Credits |
15 |
| Assessment |
- Formal examination
- Assignment
|
| Lecturer(s) |
Dr Shuo Zhou, Dr Robert Loftin |
| Resources |
Unconfirmed practical marks when available |
| Aims |
This unit aims to provide a deeper understanding of the fundamental technologies underlying data analytics at scale. In particular it will provide advanced understanding of
- parallelization of data analysis and machine learning algorithms and algorithmic approaches such as stochastic gradient descent
- practical skills relating to the deployment of scalable ML
|
| Learning Outcomes |
By the end of the unit, a student will be able to
- understand the theoretical, including mathematical, principles and wider context underpinning scalable machine learning.
- understand practical parallelization of algorithms and algorithmic approaches using such techniques as stochastic gradient descent;
- deploy a practical implementation of ML at scale, using SPARK, and programming languages such as Python/Scala;
- undertake deployment onto high performance computing platforms/clusters.
|
| Content |
Spark & HPC
- Spark overview
- High performance computing
- Spark DataFrame/dataset
- Spark machine learning pipeline
- Parallelization & optimization in Spark
Scalable logistic regression & applications
Scalable generalised linear models & applications
Scalable decision trees & applications
Scalable neural networks
Scalable matrix factorization for collaborative filtering & applications
Scalable KMeans clustering & applications
Scalable PCA for dimensionality reduction & applications
Other topics - e.g., cloud computing for Machine Learning workflows
|
| Restrictions |
Optional modules within the school have limited capacity. We will always try to accommodate all students but cannot guarantee a place. |
| Teaching Method |
Lectures, laboratory classes. |
| Feedback |
Immediately for exercises in laboratory classes. After each coursework stage through debriefing lecture and individual marking. |
|
