Module Code: |
H7AML |
Long Title
|
Advanced Machine Learning
|
Title
|
Advanced Machine Learning
|
Module Level: |
LEVEL 7 |
EQF Level: |
6 |
EHEA Level: |
First Cycle |
Module Coordinator: |
Arghir Moldovan |
Module Author: |
Arghir Moldovan |
Departments: |
School of Computing
|
Specifications of the qualifications and experience required of staff |
MSc and/or PhD degree in computer science or cognate discipline. May have industry experience also.
|
Learning Outcomes |
On successful completion of this module the learner will be able to: |
# |
Learning Outcome Description |
LO1 |
Apply and evaluate the efficacy of advanced data preparation methods |
LO2 |
Build and apply advanced methods for prediction and forecasting in various problem domains |
LO3 |
Build and evaluate advanced machine learning models in various problem domains |
LO4 |
Extract, interpret and analyse information and knowledge from non-trivial real-world datasets |
LO5 |
Summarise, critique and present the results of advanced machine learning models in various problem domains |
Dependencies |
Module Recommendations
This is prior learning (or a practical skill) that is required before enrolment on this module. While the prior learning is expressed as named NCI module(s) it also allows for learning (in another module or modules) which is equivalent to the learning specified in the named module(s).
|
67243 |
H6DMML |
Data Mining and Machine Learning |
Co-requisite Modules
|
No Co-requisite modules listed |
Entry requirements |
learners should have attained the knowledge, skills and competence gained from stage 2 of the BSc (Hons) in Data Science
|
Module Content & Assessment
Indicative Content |
General Strategies Revisited
Increasing data complexity and size with fundamental methods. Considerations of Complexity on Computing Requirements
|
General Strategies Revisited
Dimensionality Reduction (PCA, MCA, etc.). Feature Engineering. Measuring Predictor Importance
|
General Strategies Revisited
Understanding, Detecting and Handling (massive) class imbalance. Understanding Factors that can Affect Model Performance; e.g. Type III errors, selection bias, measurement errors, improper variable encoding. Ethically assessing biases.
|
Advanced Regression Models
Regression revision, and penalised models
|
Advanced Regression Models
Generalised Linear Modelling. Automated Linear Modelling via Bagging and Boosting
|
Ensembles
Ensembles:. Random Forest. Voting. Stacking. Bagging and Boosting Methods (e.g. XGBoost, AdaBoost, CART aggregation etc.)
|
Black Box Methods
Support Vector Machines and Support Vector Regression
|
Black Box Methods
Neural Networks: Classic Topologies and Activation Functions; Back Propagation; Gradient Descent and Stochastic Gradient Descent; Hyperparameter Optimisation techniques
|
Deep Regression Models
A brief introduction to deep learning applied to regression problems (e.g. GLMNet). Special emphasis to be played on when these methods are(n’t) appropriate (e.g. data volumes required).
|
Assessment Breakdown | % |
Coursework | 100.00% |
AssessmentsFull Time
Coursework |
Assessment Type: |
Continuous Assessment |
% of total: |
Non-Marked |
Assessment Date: |
n/a |
Outcome addressed: |
2,3 |
Non-Marked: |
Yes |
Assessment Description: Ongoing feedback on ongoing lab activities. |
|
Assessment Type: |
Project |
% of total: |
50 |
Assessment Date: |
n/a |
Outcome addressed: |
3,4 |
Non-Marked: |
No |
Assessment Description: Team project; applying methods of the module to real world datasets such as Kaggle, Dublinked.ie etc. |
|
Assessment Type: |
Easter Examination |
% of total: |
50 |
Assessment Date: |
n/a |
Outcome addressed: |
1,2,5 |
Non-Marked: |
No |
Assessment Description: Individual Hackathon where learners identify the fit and/or appropriateness of a variety of methods to one or more appropriately sized datasets. |
|
No End of Module Assessment |
Reassessment Requirement |
Repeat examination
Reassessment of this module will consist of a repeat examination. It is possible that there will also be a requirement to be reassessed in a coursework element.
|
Reassessment Description The repeat strategy for this module is a terminal assessment. Students will be afforded an opportunity to repeat the assessment at specified times throughout the year and all learning outcomes will be assessed in the repeat assessment.
|
NCIRL reserves the right to alter the nature and timings of assessment
Module Workload
Module Target Workload Hours 0 Hours |
Workload: Full Time |
Workload Type |
Workload Description |
Hours |
Frequency |
Average Weekly Learner Workload |
Lecture |
Classroom & Demonstrations (hours) |
24 |
Per Semester |
2.00 |
Tutorial |
Other hours (Practical/Tutorial) |
24 |
Per Semester |
2.00 |
Independent Learning |
Independent learning (hours) |
202 |
Per Semester |
16.83 |
Total Weekly Contact Hours |
4.00 |
Module Resources
Recommended Book Resources |
---|
-
Hastie, T., Tibshirani, R. & Friedman, J.. (2016), The Elements of Statistical Learning: Data Mining, Inference, and Prediction (2nd ed), Springer Series in Statistics.
-
James, G., Witten, D., Hastie, T. & Tibshirani, R.. (2017), An Introduction to Statistical Learning: with Applications in R, Springer Texts in Statistics.
-
Kuhn, M. & Johnson, K.. (2013), Applied Predictive Modeling, Springer.
-
Shalev-Shwartz, S. & Ben-David, S.. (2014), Understanding Machine Learning: From Theory to Algorithms, Cambridge University Press.
| Supplementary Book Resources |
---|
-
Downey, B.. (2014), Think Stats: Exploratory Data Analysis, (2nd ed).
-
Goodfellow, I., Bengio, Y., & Courville, A.. (2016), Deep Learning, The MIT Press.
-
Hearty, J.. (2016), Advanced Machine Learning with Python, Packt Publishing Ltd.
-
Leskovec, J. Rajaraman, A., & Ullman, J.. (2014), Mining of Massive Datasets, Cambridge University Press.
-
Wickham, H. & Grolemund, G.. (2017), R for Data Science: Import, Tidy, Transform, Visualize, and Model Data, O'Reilly.
| This module does not have any article/paper resources |
---|
Other Resources |
---|
|
|