EEEM066 Fundamentals of Machine Learning Coursework Assignment, University of Surrey

Objectives

1. Learn and become familiar with the process of sequential hyperparameter tuning of an entire deep learning pipeline.

2. Develop intuition via experimentation on how to use such model design and parameter tuning to improve the performance of a pipeline.

3. Interpret experimental results of deep learning experiments and their implications.

Outcomes

1. Navigate general deep learning codebases and modify according to requirements.

2. Empirically suggest reasonable values for hyperparameters and select the ones enabling the best performance.

3. Understand the implications of changing the backbone architecture of a pipeline.

4. Reason why data augmentation works.

5. Understand the implications and importance of selecting the best learning rate and batch size.

6. Understand how to report experimental results professionally. This includes using tables, plots, and diagrams to communicate ideas and results.

Guidelines

1. Submit your assignment as a Word or PDF document, along with the supporting  evidence as required, via the Surrey. Learn submission page before the deadline. Do not submit any other unrelated documents that could flood your submission and mislead the marking process. Only the latest version submitted before the deadline will be assessed if multiple versions exist.

2. Label your submission file as: EEEM066-[insert your URN number]-Assignment e.g., EEEM066-1234567-Assignment

3. Include a reference list at the end of your assignment, and use the Harvard or IEEE referencing style when citing sources. Please use the selected option consistently.

4. Word limit: Each section of the report should not exceed 200 words (excluding tables, plots, and graphs). Penalties of up to 50% may be applied for unnecessarily long reports.

5. This assignment requires time and effort. It is recommended that you start working on it early to minimise the possibility of experiencing stress around it and avoid missing the deadline.

6. You will receive both an overall mark and written feedback on your assignment.

7. Extensions for this assignment will only be granted under exceptional circumstances. Extension requests must be submitted via the Extenuating Circumstances (ECs) process, as tutors are not permitted to approve ad hoc extension requests.