AI Icon Artificial Intelligence

Course Details Image

Limited Time Offer

Enrol now and save $0 on your course fee

03 Days 03 Hours 03 Minutes 03 Seconds

This 2-day immersive course moves beyond the "black box" approach to machine learning. While covering core algorithms (Supervised, Unsupervised, and Reinforcement Learning), this curriculum emphasizes the complete ML lifecycle. Participants will learn not just how to train a model, but how to prepare "messy" real-world data, select the mathematically appropriate algorithms, evaluate models using advanced metrics (F1, ROC), and understand the ethical and strategic implications of deployment.

Learning Outcomes:

  • Perform Feature Engineering and data cleaning on raw datasets (Addressing Methodological Gap).
  • Implement and distinguish between Supervised (Regression/Classification) and
  • Unsupervised Learning.
  • Apply Model Selection Theory to choose algorithms based on data characteristics rather than trial-and-error.
  • Evaluate models using Precision, Recall, F1 Score, and ROC Curves.
  • Audit models for Algorithmic Bias and understand the MLOps lifecycle (Addressing Contextual Gap).

Module 1: The Reality of ML

  • Overview of ML applications and the role of AI Engineering
  • The AI Lifecycle: Moving beyond "code" to scoping and strategy
  • Gap Fill: Defining the business problem before coding (for Project Managers)

Module 2: Data Engineering & Preparation

  • The "Data-Ready" Workshop: Handling missing values, outlier detection, and normalization
  • Feature Selection vs. Feature Engineering
  • Activity: Cleaning a "broken" dataset using Pandas.

Module 3: Supervised Learning (Regression)

  • Linear Regression mechanics.
  • Theory Deep Dive: Overfitting, Underfitting, and Bias-Variance Tradeoff.
  • Case Study: Predicting house prices (Coding + Residual Analysis).

Module 4: Supervised Learning (Classification)

  • Logistic Regression & Support Vector Machines (SVM).
  • Decision Trees & Random Forests.
  • Case Study: Predicting loan approvals and classifying handwritten digits.
  • Critical Context: Why we choose Random Forest over SVM (interpretability vs. accuracy).

Challenge: The "Glass Box" Approach

  • Hands-on: Build a classifier for employee attrition
  • Hyperparameter Tuning: Using GridSearch to optimize models (moving beyond default settings).
  • Interpretability: Visualising Decision Trees to explain why a prediction was made.

Module 5: Unsupervised & Reinforcement Learning

  • Unsupervised: Clustering (K-Means) and Dimensionality Reduction (PCA).
*Important Note : Fees are subject to Singapore's prevailing Goods and Services Tax (GST).
Course Details Image
[Course Title]
Available Course Dates
Fees Enquire Now

Explore Other Courses

We couldn’t find any result
based on your selection.
Please wait a moment
while we retrieve the data

Have Question?

We’re here to help — reach out anytime.

By submitting this form, you consent to be contacted via email and/or your mobile number regarding your enquiry. You consent to the collection, use, disclosure and processing of your personal data in accordance with our Personal Data Policy.