- Week 01 - Introduction to Machine Learning
- Week 02 - Data Preprocessing
- Week 03 - Supervised Learning Methods
- Week 04 - Computer Vision
- Week 05 - Reinforcement Learning
- Week 06 - Natural Language Processing
Students must develop two Machine Learning projects during the course.
The final deliverables are:
- Research Article (7–10 pages) following an academic paper style (IEEE).
- Final Presentation (Week 8) with live Q&A.
The project will be graded in stages, each with a defined weight.
The document must include the following:
-
(20%) Literature Review & State of the Art
Synthesize high-quality sources (recent 5–10 years + seminal works). Compare and contrast methods and findings, identify convergences, contradictions, and explicit research gaps your work will address. Use consistent academic citations. -
(15%) Research Question & Objectives
Clearly state the research question and motivation grounded in the identified gap. Define SMART objectives (specific, measurable, achievable, relevant, time-bound) aligned with the project scope. -
(15%) Data & Preliminary Analysis
Describe datasets, features (X), target (y), sampling, size, and data provenance. Include basic EDA (distributions, correlations), target balance, leakage checks, and justify the evaluation metric with respect to the objective. -
(15%) Materials & Methods
Detail preprocessing steps, feature engineering, model(s) and baselines, train/validation/test strategy, hyperparameter search and validation protocol (e.g., cross-validation), and reproducibility elements (random seeds, environment). Justify design choices. -
(15%) Results
Report quantitative results with appropriate metrics, tables/figures, and (when applicable) variability estimates (e.g., std/CI). Include ablations or comparisons against baselines to show incremental value. -
(10%) Discussion
Interpret results in context of the literature and objectives. Provide error analysis, limitations, threats to validity, and implications for deployment or future research. -
(10%) Conclusions
Concise takeaways that answer the research question, reflect on objective fulfillment, and outline concrete future work.
Each team will present their project in the last class (10 minutes + 3 minutes Q&A).
-
(20%) Time Management
Deliver the presentation within the assigned time, focusing on the most relevant aspects. -
(20%) Language and Precision
Use clear and technically accurate language when describing ML methods and results. -
(20%) Methodology Explanation
Present the workflow and reasoning in a structured, logical order. -
(20%) Supporting Material
Slides must support the talk (not replace it), highlighting results with appropriate visuals. -
(20%) Handling Questions
Respond with confidence, technical rigor, and respect to the audience’s questions.
- Written Report (Research Article): 70%
- Final Presentation: 30%
