Module 3: Pros & Cons

3.1 Advantages of Machine Learning

Machine Learning provides a range of benefits that make it a powerful tool for solving complex problems, automating processes, and making intelligent decisions based on data.

Example: Netflix uses ML to automatically recommend shows that a user is likely to enjoy, increasing user engagement and retention.
Key Advantages:
  • Automation: Reduces the need for manual intervention in repetitive tasks.
  • Efficiency: Quickly processes large volumes of data and detects patterns.
  • Adaptability: Improves performance over time as more data is fed.
  • Personalization: Enables customized experiences in apps, ads, and services.
  • Scalability: Works across industries and datasets of varying sizes.
Did You Know?
  • Google's search algorithm and autocomplete suggestions are heavily powered by ML.
  • ML can outperform doctors in detecting certain diseases from X-rays and MRIs.

3.2 Disadvantages of Machine Learning

Despite its power, Machine Learning comes with certain limitations and challenges that must be addressed to avoid pitfalls and ensure ethical use.

Example: An ML hiring system trained on biased historical data might learn to favor certain groups unfairly, leading to discrimination.
Key Disadvantages:
  • Data Dependency: ML models require large, clean, and relevant datasets to perform well.
  • Lack of Transparency: Deep learning models often operate as "black boxes" with limited interpretability.
  • Bias & Fairness: Biased data leads to biased predictions, raising ethical concerns.
  • High Resource Costs: Training models (especially deep learning) can require significant computing power and energy.
  • Security Risks: Models can be attacked or manipulated (e.g., adversarial inputs).
Did You Know?
  • MIT researchers tricked an image classifier into identifying a turtle as a rifle — simply by tweaking a few pixels.
  • Some facial recognition systems perform poorly on underrepresented ethnic groups, raising major privacy and fairness issues.