Universitat Internacional de Catalunya
Artificial Intelligence
Other languages of instruction: Catalan, Spanish,
Teaching staff
Send an email to set up a meeting to:
Marius Khan:
Introduction
This course offers a comprehensive understanding of how artificial intelligence (AI) and machine learning are revolutionizing the cosmetic industry. AI has become a powerful tool for innovation, offering cutting-edge approaches for advanced formulations in product discovery and development, cosmetic procedures, and even personalized treatments and recommendations for each person’s unique needs. The application of AI in the sector will boost cosmetics effectiveness, save production costs, and minimize environmental impact by adjusting waste or non-used products. Students will learn to harness AI for creating innovative cosmetic formulations, and discover how AI is transforming everything from product development to personalized skincare solutions.
Objectives
- To understand the basics of artificial intelligence (AI) and machine learning (ML)
- To know specific AI/ML technology that is applicable in cosmetics => NLP, Reinforcement Learning, Deep Learning, Clustering and Classification Algorithms, etc.
- To familiarize with programming libraries, frameworks and software landscapes necessary to get started with first small data driven projects.
- To know current unmet needs and challenges in cosmetics industry and value propositions of AI/ML.
To understand where AI/ML would give more value to specific use cases in the cosmetics industry than traditional programming.
Competences/Learning outcomes of the degree programme
CE3 - To have the ability to understand and use methodologies, new technologies and bioengineering tools in the research, development and manufacturing of cosmetic products.
CE6 - To integrate the fundamentals of materials science and technology, nanotechnology and 3D printing for their application in modern cosmetics, in addition to knowing how to apply artificial intelligence tools for the design of new cosmetic products.
CE9 - To promote the spirit of entrepreneurship and integrate the knowledge applied to the organization and management of companies in Bioengineering, considering their legal framework and current regulations, as well as the process necessary to preserve the intellectual property of cosmetic products.
CE11 - To have the ability to carry out a project using data sources, and the application of methodologies, research techniques and tools specific to Bioengineering, and make a public presentation and defense of the project before a specialized audience in such a way as to demonstrate the acquisition of the skills and knowledge specific to the master's degree.
Learning outcomes of the subject
Finalizing this course, the student will be able to:
- Understand on a technical level what Artificial Intelligence is and what methods exists like machine learning and how to distinguish between them
- Challenges to realize AI systems
- Understand the limitations of AI and machine learnings systems
- What kind of tools/frameworks exists to build own Machine Learning systems
- What kind of software landscape is necessary to get out of the lab-ready situation into production readiness.
In which areas AI can be applied in the field of cosmetics
RA2 - Understand and use bioengineering tools, such as 3D printing, artificial intelligence, or the use of materials, for innovation in the research, development and manufacture of cosmetic products.
Syllabus
1. Introduction of Artificial Intelligence, Machine Learning, Deep Learning and Neural Networks
Understanding AI, ML and Deep Learning.
Clarify confusions in terminology and general misunderstandings.
Key Machine Learning Paradigms: Supervised, unsupervised Learning. Transfer Learning.
Brief History of AI/ML in general.
Brief History of AI in dermatology with the biggest milestones so far.
2. Specific AI/ML technology types applicable in cosmetics
Natural Language Processing (NLP)
Reinforcement Learning
Deep Learning (Neural Networks)
Clustering and Classification
General Adversal Networks (GANs)
3. Software Frameworks and Libraries for AI/ML – necessary software enterprise landscape for AI driven projects
4. Specific Use cases of AI/ML in cosmetics
Example of market players
Formulation optimization
Customer satisfaction
Quality control
5. Future AI/ML trends and how it would impact the cosmetics industry
6. Limitations of AI/ML systems
Challenges of realizing AI Systems.
When does it make sense to realize AI systems? AI vs. traditional programming
Teaching and learning activities
In person
The course will be divided mainly into master classes, problems and group project sessions and computing sessions.
Classes will be taught in English. The didactic material will be presented mainly in English.
Eventually, the teacher could use the Moodle platform that could include various resources, such as forms, exercises, multimedia material ... that the student must complete the subject.
The list of ECTS credits and the workload in learning hours depend on the different methodologies that will be used. Each ECTS theoretical credit has 10 hours in which the teacher has a presence in the classroom. The rest of the hours up to 25 correspond to the load of directed and autonomous learning of the student. This last teaching load can be done through autonomous activities, group work that will be presented and defended in class or individual study necessary to achieve the learning objectives of the different subjects.
Evaluation systems and criteria
In person
Throughout the semester, students will work in teams to develop their own AI Beauty Platform, moving step by step from idea generation to a clickable prototype. The project is divided into three main phases, each with specific deliverables and evaluation components.
Phase 1 – Market Research & Concept Development (25%)
- Concept Paper (20%): Students research the current state of the art in AI-powered beauty platforms, identify unmet needs and market gaps, and propose a clear concept for their platform. The paper should highlight the unique value proposition and vision.
- Pitch Presentation (5%): Short, persuasive pitch of the concept supported by a concise slide deck.
Phase 2 – Business Idea & Strategic Planning (30%)
- Business Plan (25%): Students expand their concept into a detailed business idea, specifying how AI/ML technologies will be integrated into the beauty platform, outlining strategies for data acquisition, and presenting a realistic path to implementation (without going into deep technical details).
- Pitch Presentation (5%): Professional presentation of the business strategy, emphasizing the platform’s innovative potential and feasibility.
Phase 3 – Prototyping & Mockups (45%)
- Mockups / Clickable Prototype (40%): Using tools such as Figma, students create detailed mockups and a clickable prototype of their AI Beauty Platform. The prototype should demonstrate core functionality, usability, and alignment with the earlier concept and business plan.
- Pitch Presentation (5%): Live demonstration of the prototype with focus on user experience, design quality, and persuasiveness of the solution.
General Criteria Across All Phases:
- Depth and quality of research.
- Originality, creativity, and innovation in the AI Beauty Platform idea.
- Practical relevance and market fit.
- Clear structure, professionalism, and persuasiveness in both written and oral outputs.
- Team collaboration and effective project management.
Students must achieve at least 5.0/10 in each project phase (Phase 1, Phase 2, and Phase 3). A final weighted grade of 5.0/10 or higher is required to pass the course.
If the minimum grade is not reached in any one phase, the project will be considered incomplete and the course cannot be passed, regardless of the overall average.
Active participation in teamwork and presentations is mandatory; insufficient contribution may lower the individual mark.
Attendance is mandatory for all workgroup projects and problem sessions and must be higher than 90% to pass the course.
Assignments will not be accepted after the deadline and will only be accepted by Moodle and not by email, with the exception of the in-class problems/projects that only will be allowed during the session.
Important considerations
- Plagiarism, copying, or any other action that may be considered cheating will be zero in that evaluation section. Besides, in exams, it will mean the immediate failure of the whole subject.
- In the second-sitting exams, the maximum grade students will be able to obtain is "Excellent" (grade with honors distinction will not be possible).
- Changes in the calendar, exam dates, or the evaluation system will not be accepted.
- Exchange students (Erasmus and others) or repeaters will be subjected to the same conditions as the rest of the students
Bibliography and resources
Géron, A. (2019). Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems (2nd ed.). O’Reilly Media.