Study Guide

Objective

The student is able to carry out a development project that includes practical machine learning. The student is able to apply appropriate modern tools and methods to development work. The student is able to make decisions in unforeseen situations and utilize various communication channels and tools in the external and internal communication of the project. The student is capable of multicultural collaboration.

Content

- A multicultural work community
- Automation Testing / DevOps

Location and time

Classes and workshops take place on the Rovaniemi Jokiväylä campus in the scheduled classroom. Additional team meetings will be arranged as needed.

Materials

Selected research articles and online resources provided throughout the course.

Reading:

Reinforcement Learning: An Introduction, Richard S. Sutton and Andrew G. Barto. Second Edition, MIT Press, Cambridge, MA, 2018

Link: http://incompleteideas.net/book/the-book-2nd.html

Deep Learning (Adaptive Computation and Machine Learning series), An MIT Press book, Ian Goodfellow and Yoshua Bengio and Aaron Courville

Link: https://www.deeplearningbook.org/

Teaching methods

The study module employs project-based learning, where students collaborate in teams to develop practical machine learning and AI solutions utilizing DevOps principles and methods. Regular workshops, seminars, and mentoring sessions will support the project work. Active participation and engagement in all activities are expected.

Employer connections

The study module aims to include collaboration with industry partners, offering students real-world problems to solve. This provides practical experience and networking opportunities within the field.

Exam schedules

There are no traditional exams. Assessment is continuous and based on project milestones, presentations, and final deliverables. Opportunities for resubmission will be provided for components that do not meet the required standards.

International connections

International students are welcome, and all course activities are conducted in English.

Completion alternatives

Alternative completion methods may be applicable through the process of identification and recognition of acquired competencies.

Content scheduling

1. Introduction, team formation, project topic selection
2. Project planning, literature review, initial development
3. Automation and testing
4. Mid-project reviews, implementation
5. Refinement, troubleshooting, preparation for final presentation
6. Final presentations, submission of project reports

Further information

Students are expected to maintain regular communication with instructors and team members. All resources and announcements will be available on the study module's online platform (Moodle).

Evaluation scale

H-5

Assessment methods and criteria

Project Development (50%): Quality, innovation, and functionality of the final product.
Team Collaboration (20%): Contribution to team efforts and effective communication.
Presentations (15%): Clarity, professionalism, and ability to articulate project details.
Documentation (15%): Completeness, clarity, and organization of reports and code documentation.

Final Seminar, Project Reporting and Peer Reviews are essential parts of the assessment.

Grading of 0 to 5 is used where 0 is Fail and 5 is Excellent

Assessment criteria, satisfactory (1-2)

Basic machine learning and AI concepts are implemented, resulting in a product with some functionality. Contribution to team efforts is observable but may be inconsistent. Presentations cover essential aspects of the project but may lack clarity or depth. Documentation provides a basic overview but might be lacking in detail or organization.

Assessment criteria, good (3-4)

Machine learning and AI concepts are applied effectively, producing a functional product with moderate technical complexity. Active participation in team activities is demonstrated, contributing positively to the team's objectives. Presentations are clear and professional, conveying technical details adequately. Documentation is well-organized and offers a good understanding of the project's technical aspects.

Assessment criteria, excellent (5)

Advanced machine learning and AI techniques are applied proficiently, resulting in a highly functional and innovative product. Significant contributions to the team are made, potentially taking on coordination roles and promoting effective collaboration. Presentations are professional and clear, thoroughly explaining the project's technical components. Documentation is comprehensive and well-structured, providing detailed insights into the project's development and implementation.